set version to 0.74.1.dev

Correct a typo

.github/workflows/docker-build-test.yml

@@ -4,23 +4,25 @@ on:
   push:
     paths-ignore:
       - 'aider/website/**'
-      - README.md
-      - HISTORY.md
+      - 'README.md'
+      - 'HISTORY.md'
+      - '.github/workflows/*'
+      - '!.github/workflows/docker-build-test.yml'
     branches:
       - main
   pull_request:
     paths-ignore:
       - 'aider/website/**'
-      - README.md
+      - 'README.md'
+      - 'HISTORY.md'
+      - '.github/workflows/*'
+      - '!.github/workflows/docker-build-test.yml'
     branches:
       - main
 
-# copy most of these steps from release.yml, but push: false and no tags:
-
 jobs:
-  build:
+  docker_build_and_push:
     runs-on: ubuntu-latest
-
     steps:
     - name: Checkout code
       uses: actions/checkout@v4
@@ -29,11 +31,19 @@ jobs:
 
     - name: Set up QEMU
       uses: docker/setup-qemu-action@v3
-      
+
     - name: Set up Docker Buildx
       uses: docker/setup-buildx-action@v3
 
-    - name: Build Docker standard image
+    - name: Login to DockerHub
+      if: ${{ github.event_name != 'pull_request' }}
+      uses: docker/login-action@v3
+      with:
+        username: ${{ secrets.DOCKERHUB_USERNAME }}
+        password: ${{ secrets.DOCKERHUB_PASSWORD }}
+
+    - name: Build Docker images (PR)
+      if: ${{ github.event_name == 'pull_request' }}
       uses: docker/build-push-action@v5
       with:
         context: .
@@ -42,7 +52,19 @@ jobs:
         push: false
         target: aider
 
-    - name: Build Docker full image
+    - name: Build Docker images (Push)
+      if: ${{ github.event_name != 'pull_request' }}
+      uses: docker/build-push-action@v5
+      with:
+        context: .
+        file: ./docker/Dockerfile
+        platforms: linux/amd64,linux/arm64
+        push: true
+        tags: ${{ secrets.DOCKERHUB_USERNAME }}/aider:dev
+        target: aider
+
+    - name: Build Docker full image (PR)
+      if: ${{ github.event_name == 'pull_request' }}
       uses: docker/build-push-action@v5
       with:
         context: .
@@ -50,3 +72,14 @@ jobs:
         platforms: linux/amd64,linux/arm64
         push: false
         target: aider-full
+
+    - name: Build Docker full image (Push)
+      if: ${{ github.event_name != 'pull_request' }}
+      uses: docker/build-push-action@v5
+      with:
+        context: .
+        file: ./docker/Dockerfile
+        platforms: linux/amd64,linux/arm64
+        push: true
+        tags: ${{ secrets.DOCKERHUB_USERNAME }}/aider-full:dev
+        target: aider-full

.github/workflows/pages.yml

@@ -12,6 +12,7 @@ on:
       - "main"
     paths:
       - "aider/website/**"
+      - ".github/workflows/pages.yml"
 
   # Allows you to run this workflow manually from the Actions tab
   workflow_dispatch:
@@ -55,10 +56,9 @@ jobs:
         env:
           JEKYLL_ENV: production
       - name: Upload artifact
-        # Automatically uploads an artifact from the './_site' directory by default
-        uses: actions/upload-pages-artifact@v1
+        uses: actions/upload-pages-artifact@v3
         with:
-          path: "aider/website/_site/"
+          path: "aider/website/_site"
 
   # Deployment job
   deploy:
@@ -70,7 +70,7 @@ jobs:
     steps:
       - name: Deploy to GitHub Pages
         id: deployment
-        uses: actions/deploy-pages@v2
+        uses: actions/deploy-pages@v4
         
       - name: Set up Python 3.12
         uses: actions/setup-python@v5

.github/workflows/ubuntu-tests.yml

@@ -4,14 +4,19 @@ on:
   push:
     paths-ignore:
       - 'aider/website/**'
-      - README.md
-      - HISTORY.md
+      - 'README.md'
+      - 'HISTORY.md'
+      - '.github/workflows/*'
+      - '!.github/workflows/ubuntu-tests.yml'
     branches:
       - main
   pull_request:
     paths-ignore:
       - 'aider/website/**'
-      - README.md
+      - 'README.md'
+      - 'HISTORY.md'
+      - '.github/workflows/*'
+      - '!.github/workflows/ubuntu-tests.yml'
     branches:
       - main
 

.github/workflows/windows-tests.yml

@@ -4,14 +4,19 @@ on:
   push:
     paths-ignore:
       - 'aider/website/**'
-      - README.md
-      - HISTORY.md
+      - 'README.md'
+      - 'HISTORY.md'
+      - '.github/workflows/*'
+      - '!.github/workflows/windows-tests.yml'
     branches:
       - main
   pull_request:
     paths-ignore:
       - 'aider/website/**'
-      - README.md
+      - 'README.md'
+      - 'HISTORY.md'
+      - '.github/workflows/*'
+      - '!.github/workflows/windows-tests.yml'
     branches:
       - main
 

.gitignore

@@ -15,3 +15,4 @@ aider/_version.py
 .venv/
 .#*
 .gitattributes
+tmp.benchmarks/

CONTRIBUTING.md

@@ -56,13 +56,6 @@ It is recommended to create a virtual environment outside of the repository to k
 python -m venv /path/to/venv
 ```
 
-#### Using `virtualenv` (for older Python versions)
-
-```
-pip install virtualenv
-virtualenv /path/to/venv
-```
-
 ### Activate the Virtual Environment
 
 #### On Windows

HISTORY.md

@@ -1,6 +1,124 @@
 # Release history
 
-### main branch
+### Aider v0.74.0
+
+- Dynamically changes the Ollama context window to hold the current chat.
+- Better support for o3-mini, DeepSeek V3 & R1, o1-mini, o1 via secondary API providers.
+- Remove `<think>` tags from R1 responses for commit messages (and other weak model uses).
+- Can now specify `use_temperature: <float>` in model settings, not just true/false.
+- The full docker container now includes `boto3` for Bedrock.
+- Docker containers now set `HOME=/app` which is the normal project mount-point, to persist `~/.aider`.
+- Bugfix to prevent creating incorrect filenames like `python`, `php`, etc.
+- Bugfix for `--timeout`
+- Bugfix so that `/model` now correctly reports that the weak model is not changed.
+- Bugfix so that multi-line mode persists through ^C at confirmation prompts.
+- Watch files now fully ignores top-level directories named in ignore files, to reduce the chance of hitting OS watch limits. Helpful to ignore giant subtrees like `node_modules`.
+- Fast startup with more providers and when model metadata provided in local files.
+- Improved .gitignore handling:
+  - Honor ignores already in effect regardless of how they've been configured.
+  - Check for .env only when the file exists.
+- Yes/No prompts now accept All/Skip as alias for Y/N even when not processing a group of confirmations.
+- Aider wrote 77% of the code in this release.
+
+### Aider v0.73.0
+
+- Full support for o3-mini: `aider --model o3-mini`
+- New `--reasoning-effort` argument: low, medium, high.
+- Improved handling of context window size limits, with better messaging and Ollama-specific guidance.
+- Added support for removing model-specific reasoning tags from responses with `remove_reasoning: tagname` model setting.
+- Auto-create parent directories when creating new files, by xqyz.
+- Support for R1 free on OpenRouter: `--model openrouter/deepseek/deepseek-r1:free`
+- Aider wrote 69% of the code in this release.
+
+### Aider v0.72.3
+
+- Enforce user/assistant turn order to avoid R1 errors, by miradnanali.
+- Case-insensitive model name matching while preserving original case.
+
+### Aider v0.72.2
+- Harden against user/assistant turn order problems which cause R1 errors.
+
+### Aider v0.72.1
+- Fix model metadata for `openrouter/deepseek/deepseek-r1`
+
+### Aider v0.72.0
+- Support for DeepSeek R1.
+  - Use shortcut: `--model r1`
+  - Also via OpenRouter: `--model openrouter/deepseek/deepseek-r1`
+- Added Kotlin syntax support to repo map, by Paul Walker.
+- Added `--line-endings` for file writing, by Titusz Pan.
+- Added examples_as_sys_msg=True for GPT-4o models, improves benchmark scores.
+- Bumped all dependencies, to pick up litellm support for o1 system messages.
+- Bugfix for turn taking when reflecting lint/test errors.
+- Aider wrote 52% of the code in this release.
+
+### Aider v0.71.1
+
+- Fix permissions issue in Docker images.
+- Added read-only file announcements.
+- Bugfix: ASCII fallback for unicode errors.
+- Bugfix: integer indices for list slicing in repomap calculations.
+
+### Aider v0.71.0
+
+- Prompts to help DeepSeek work better when alternating between `/ask` and `/code`.
+- Streaming pretty LLM responses is smoother and faster for long replies.
+- Streaming automatically turns of for model that don't support it
+  - Can now switch to/from `/model o1` and a streaming model
+- Pretty output remains enabled even when editing files with triple-backtick fences
+- Bare `/ask`, `/code` and `/architect` commands now switch the chat mode.
+- Increased default size of the repomap.
+- Increased max chat history tokens limit from 4k to 8k.
+- Turn off fancy input and watch files if terminal is dumb.
+- Added support for custom voice format and input device settings.
+- Disabled Streamlit email prompt, by apaz-cli.
+- Docker container runs as non-root user.
+- Fixed lint command handling of nested spaced strings, by Aaron Weisberg.
+- Added token count feedback when adding command output to chat.
+- Improved error handling for large audio files with automatic format conversion.
+- Improved handling of git repo index errors, by Krazer.
+- Improved unicode handling in console output with ASCII fallback.
+- Added AssertionError, AttributeError to git error handling.
+- Aider wrote 60% of the code in this release.
+
+### Aider v0.70.0
+
+- Full support for o1 models.
+- Watch files now honors `--subtree-only`, and only watches that subtree.
+- Improved prompting for watch files, to work more reliably with more models.
+- New install methods via uv, including one-liners.
+- Support for openrouter/deepseek/deepseek-chat model.
+- Better error handling when interactive commands are attempted via `/load` or `--load`.
+- Display read-only files with abs path if its shorter than rel path.
+- Ask 10% of users to opt-in to analytics.
+- Bugfix for auto-suggest.
+- Gracefully handle unicode errors in git path names.
+- Aider wrote 74% of the code in this release.
+
+### Aider v0.69.1
+
+- Fix for gemini model names in model metadata.
+- Show hints about AI! and AI? when user makes AI comments.
+- Support for running without git installed.
+- Improved environment variable setup messages on Windows.
+
+### Aider v0.69.0
+
+- [Watch files](https://aider.chat/docs/usage/watch.html) improvements:
+  - Use `# ... AI?` comments to trigger aider and ask questions about your code.
+  - Now watches *all* files, not just certain source files.
+  - Use `# AI comments`, `// AI comments`, or `-- AI comments` to give aider instructions in any text file.
+- Full support for Gemini Flash 2.0 Exp:
+  - `aider --model flash` or `aider --model gemini/gemini-2.0-flash-exp`
+- [New `--multiline` flag and `/multiline-mode` command](https://aider.chat/docs/usage/commands.html#entering-multi-line-chat-messages) makes ENTER a soft newline and META-ENTER send the message, by @miradnanali.
+- `/copy-context <instructions>` now takes optional "instructions" when [copying code context to the clipboard](https://aider.chat/docs/usage/copypaste.html#copy-aiders-code-context-to-your-clipboard-paste-into-the-web-ui).
+- Improved clipboard error handling with helpful requirements install info.
+- Ask 5% of users if they want to opt-in to analytics.
+- `/voice` now lets you edit the transcribed text before sending.
+- Disabled auto-complete in Y/N prompts.
+- Aider wrote 68% of the code in this release.
+
+### Aider v0.68.0
 
 - [Aider works with LLM web chat UIs](https://aider.chat/docs/usage/copypaste.html).
   - New `--copy-paste` mode.

MANIFEST.in

@@ -0,0 +1,20 @@
+# This needs to sync with aider/help_pats.py
+
+global-exclude .DS_Store
+
+recursive-exclude aider/website/examples *
+recursive-exclude aider/website/_posts *
+
+exclude aider/website/HISTORY.md
+exclude aider/website/docs/benchmarks*.md
+exclude aider/website/docs/ctags.md
+exclude aider/website/docs/unified-diffs.md
+
+exclude aider/website/install.ps1
+exclude aider/website/install.sh
+
+recursive-exclude aider/website/docs/leaderboards *
+recursive-exclude aider/website/assets *
+recursive-exclude aider/website *.js
+recursive-exclude aider/website *.html
+recursive-exclude aider/website *.yml

README.md

@@ -5,9 +5,9 @@
 
 Aider lets you pair program with LLMs,
 to edit code in your local git repository.
-Start a new project or work with an existing git repo.
-Aider works best with GPT-4o & Claude 3.5 Sonnet and can 
-[connect to almost any LLM](https://aider.chat/docs/llms.html).
+Start a new project or work with an existing code base.
+Aider works best with Claude 3.5 Sonnet, DeepSeek V3, o1 & GPT-4o and can [connect to almost any LLM](https://aider.chat/docs/llms.html).
+
 
 <!-- SCREENCAST START -->
 <p align="center">
@@ -43,28 +43,36 @@ VIDEO END -->
 cog.out(open("aider/website/_includes/get-started.md").read())
 ]]]-->
 
-You can get started quickly like this:
+If you already have python 3.8-3.13 installed, you can get started quickly like this:
 
-```
-python -m pip install -U aider-chat
+```bash
+python -m pip install aider-install
+aider-install
 
-# Change directory into a git repo
-cd /to/your/git/repo
+# Change directory into your code base
+cd /to/your/project
 
-# Work with Claude 3.5 Sonnet on your repo
-export ANTHROPIC_API_KEY=your-key-goes-here
-aider
+# Work with DeepSeek via DeepSeek's API
+aider --model deepseek --api-key deepseek=your-key-goes-here
 
-# Work with GPT-4o on your repo
-export OPENAI_API_KEY=your-key-goes-here
-aider 
+# Work with Claude 3.5 Sonnet via Anthropic's API
+aider --model sonnet --api-key anthropic=your-key-goes-here
+
+# Work with GPT-4o via OpenAI's API
+aider --model gpt-4o --api-key openai=your-key-goes-here
+
+# Work with Sonnet via OpenRouter's API
+aider --model openrouter/anthropic/claude-3.5-sonnet --api-key openrouter=your-key-goes-here
+
+# Work with DeepSeek via OpenRouter's API
+aider --model openrouter/deepseek/deepseek-chat --api-key openrouter=your-key-goes-here
 ```
 <!--[[[end]]]-->
 
 See the
 [installation instructions](https://aider.chat/docs/install.html)
-and other
-[documentation](https://aider.chat/docs/usage.html)
+and
+[usage documentation](https://aider.chat/docs/usage.html)
 for more details.
 
 ## Features
@@ -73,21 +81,22 @@ for more details.
 - Ask for changes:
   - Add new features or test cases.
   - Describe a bug.
-  - Paste in an error message or or GitHub issue URL.
+  - Paste in an error message or GitHub issue URL.
   - Refactor code.
   - Update docs.
 - Aider will edit your files to complete your request.
 - Aider [automatically git commits](https://aider.chat/docs/git.html) changes with a sensible commit message.
+- [Use aider inside your favorite editor or IDE](https://aider.chat/docs/usage/watch.html).
 - Aider works with [most popular languages](https://aider.chat/docs/languages.html): python, javascript, typescript, php, html, css, and more...
-- Aider works best with GPT-4o & Claude 3.5 Sonnet and can [connect to almost any LLM](https://aider.chat/docs/llms.html).
 - Aider can edit multiple files at once for complex requests.
 - Aider uses a [map of your entire git repo](https://aider.chat/docs/repomap.html), which helps it work well in larger codebases.
-- Edit files in your editor while chatting with aider,
+- Edit files in your editor or IDE while chatting with aider,
 and it will always use the latest version.
 Pair program with AI.
 - [Add images to the chat](https://aider.chat/docs/usage/images-urls.html) (GPT-4o, Claude 3.5 Sonnet, etc).
 - [Add URLs to the chat](https://aider.chat/docs/usage/images-urls.html) and aider will read their content.
 - [Code with your voice](https://aider.chat/docs/usage/voice.html).
+- Aider works best with Claude 3.5 Sonnet, DeepSeek V3, o1 & GPT-4o and can [connect to almost any LLM](https://aider.chat/docs/llms.html).
 
 
 ## Top tier performance

aider/__init__.py

@@ -1,6 +1,6 @@
 from packaging import version
 
-__version__ = "0.68.1.dev"
+__version__ = "0.74.1.dev"
 safe_version = __version__
 
 try:

aider/analytics.py

@@ -12,6 +12,46 @@ from aider import __version__
 from aider.dump import dump  # noqa: F401
 from aider.models import model_info_manager
 
+PERCENT = 10
+
+
+def compute_hex_threshold(percent):
+    """Convert percentage to 6-digit hex threshold.
+
+    Args:
+        percent: Percentage threshold (0-100)
+
+    Returns:
+        str: 6-digit hex threshold
+    """
+    return format(int(0xFFFFFF * percent / 100), "06x")
+
+
+def is_uuid_in_percentage(uuid_str, percent):
+    """Check if a UUID string falls within the first X percent of the UUID space.
+
+    Args:
+        uuid_str: UUID string to test
+        percent: Percentage threshold (0-100)
+
+    Returns:
+        bool: True if UUID falls within the first X percent
+    """
+    if not (0 <= percent <= 100):
+        raise ValueError("Percentage must be between 0 and 100")
+
+    if not uuid_str:
+        return False
+
+    # Convert percentage to hex threshold (1% = "04...", 10% = "1a...", etc)
+    # Using first 6 hex digits
+    if percent == 0:
+        return False
+
+    threshold = compute_hex_threshold(percent)
+    return uuid_str[:6] <= threshold
+
+
 mixpanel_project_token = "6da9a43058a5d1b9f3353153921fb04d"
 posthog_project_api_key = "phc_99T7muzafUMMZX15H8XePbMSreEUzahHbtWjy3l5Qbv"
 posthog_host = "https://us.i.posthog.com"
@@ -84,31 +124,7 @@ class Analytics:
         if not self.user_id:
             return False
 
-        PERCENT = 2.5
-        return self.is_uuid_in_percentage(self.user_id, PERCENT)
-
-    def is_uuid_in_percentage(self, uuid_str, percent):
-        """Check if a UUID string falls within the first X percent of the UUID space.
-
-        Args:
-            uuid_str: UUID string to test
-            percent: Percentage threshold (0-100)
-
-        Returns:
-            bool: True if UUID falls within the first X percent
-        """
-        if not (0 <= percent <= 100):
-            raise ValueError("Percentage must be between 0 and 100")
-
-        if not uuid_str:
-            return False
-
-        # Convert percentage to hex threshold (1% = "04...", 10% = "1a...", etc)
-        # Using first 6 hex digits
-        if percent == 0:
-            return False
-        threshold = format(int(0xFFFFFF * percent / 100), "06x")
-        return uuid_str[:6] <= threshold
+        return is_uuid_in_percentage(self.user_id, PERCENT)
 
     def get_data_file_path(self):
         try:
@@ -181,6 +197,7 @@ class Analytics:
 
     def posthog_error(self):
         """disable posthog if we get an error"""
+        print("X" * 100)
         # https://github.com/PostHog/posthog-python/blob/9e1bb8c58afaa229da24c4fb576c08bb88a75752/posthog/consumer.py#L86
         # https://github.com/Aider-AI/aider/issues/2532
         self.ph = None
@@ -227,3 +244,7 @@ class Analytics:
                     f.write("\n")
             except OSError:
                 pass  # Ignore OS errors when writing to logfile
+
+
+if __name__ == "__main__":
+    dump(compute_hex_threshold(PERCENT))

aider/args.py

@@ -106,7 +106,7 @@ def get_parser(default_config_files, git_root):
         const=gpt_3_model_name,
         help=f"Use {gpt_3_model_name} model for the main chat",
     )
-    deepseek_model = "deepseek/deepseek-coder"
+    deepseek_model = "deepseek/deepseek-chat"
     group.add_argument(
         "--deepseek",
         action="store_const",
@@ -203,6 +203,11 @@ def get_parser(default_config_files, git_root):
         metavar="ALIAS:MODEL",
         help="Add a model alias (can be used multiple times)",
     )
+    group.add_argument(
+        "--reasoning-effort",
+        type=str,
+        help="Set the reasoning_effort API parameter (default: not set)",
+    )
     group.add_argument(
         "--verify-ssl",
         action=argparse.BooleanOptionalAction,
@@ -211,7 +216,7 @@ def get_parser(default_config_files, git_root):
     )
     group.add_argument(
         "--timeout",
-        type=int,
+        type=float,
         default=None,
         help="Timeout in seconds for API calls (default: None)",
     )
@@ -287,7 +292,7 @@ def get_parser(default_config_files, git_root):
         "--map-tokens",
         type=int,
         default=None,
-        help="Suggested number of tokens to use for repo map, use 0 to disable (default: 1024)",
+        help="Suggested number of tokens to use for repo map, use 0 to disable",
     )
     group.add_argument(
         "--map-refresh",
@@ -427,7 +432,8 @@ def get_parser(default_config_files, git_root):
         default="default",
         help=(
             "Set the markdown code theme (default: default, other options include monokai,"
-            " solarized-dark, solarized-light)"
+            " solarized-dark, solarized-light, or a Pygments builtin style,"
+            " see https://pygments.org/styles for available themes)"
         ),
     )
     group.add_argument(
@@ -765,6 +771,12 @@ def get_parser(default_config_files, git_root):
         default="utf-8",
         help="Specify the encoding for input and output (default: utf-8)",
     )
+    group.add_argument(
+        "--line-endings",
+        choices=["platform", "lf", "crlf"],
+        default="platform",
+        help="Line endings to use when writing files (default: platform)",
+    )
     group.add_argument(
         "-c",
         "--config",
@@ -795,6 +807,12 @@ def get_parser(default_config_files, git_root):
         default=True,
         help="Enable/disable fancy input with history and completion (default: True)",
     )
+    group.add_argument(
+        "--multiline",
+        action=argparse.BooleanOptionalAction,
+        default=False,
+        help="Enable/disable multi-line input mode with Meta-Enter to submit (default: False)",
+    )
     group.add_argument(
         "--detect-urls",
         action=argparse.BooleanOptionalAction,

aider/coders/ask_prompts.py

@@ -7,6 +7,8 @@ class AskPrompts(CoderPrompts):
     main_system = """Act as an expert code analyst.
 Answer questions about the supplied code.
 Always reply to the user in {language}.
+
+Describe code changes however you like. Don't use SEARCH/REPLACE blocks!
 """
 
     example_messages = []

aider/coders/base_coder.py

@@ -27,10 +27,10 @@ from aider.history import ChatSummary
 from aider.io import ConfirmGroup, InputOutput
 from aider.linter import Linter
 from aider.llm import litellm
+from aider.models import RETRY_TIMEOUT
 from aider.repo import ANY_GIT_ERROR, GitRepo
 from aider.repomap import RepoMap
 from aider.run_cmd import run_cmd
-from aider.sendchat import RETRY_TIMEOUT, send_completion
 from aider.utils import format_content, format_messages, format_tokens, is_image_file
 
 from ..dump import dump  # noqa: F401
@@ -59,7 +59,8 @@ def wrap_fence(name):
 
 
 all_fences = [
-    ("``" + "`", "``" + "`"),
+    ("`" * 3, "`" * 3),
+    ("`" * 4, "`" * 4),  # LLMs ignore and revert to triple-backtick, causing #2879
     wrap_fence("source"),
     wrap_fence("code"),
     wrap_fence("pre"),
@@ -84,7 +85,7 @@ class Coder:
     max_reflections = 3
     edit_format = None
     yield_stream = False
-    temperature = 0
+    temperature = None
     auto_lint = True
     auto_test = False
     test_cmd = None
@@ -143,7 +144,13 @@ class Coder:
             # the system prompt.
             done_messages = from_coder.done_messages
             if edit_format != from_coder.edit_format and done_messages and summarize_from_coder:
-                done_messages = from_coder.summarizer.summarize_all(done_messages)
+                try:
+                    done_messages = from_coder.summarizer.summarize_all(done_messages)
+                except ValueError:
+                    # If summarization fails, keep the original messages and warn the user
+                    io.tool_warning(
+                        "Chat history summarization failed, continuing with full history"
+                    )
 
             # Bring along context from the old Coder
             update = dict(
@@ -230,10 +237,10 @@ class Coder:
             if map_tokens > 0:
                 refresh = self.repo_map.refresh
                 lines.append(f"Repo-map: using {map_tokens} tokens, {refresh} refresh")
-                max_map_tokens = 2048
+                max_map_tokens = self.main_model.get_repo_map_tokens() * 2
                 if map_tokens > max_map_tokens:
                     lines.append(
-                        f"Warning: map-tokens > {max_map_tokens} is not recommended as too much"
+                        f"Warning: map-tokens > {max_map_tokens} is not recommended. Too much"
                         " irrelevant code can confuse LLMs."
                     )
             else:
@@ -245,9 +252,16 @@ class Coder:
         for fname in self.get_inchat_relative_files():
             lines.append(f"Added {fname} to the chat.")
 
+        for fname in self.abs_read_only_fnames:
+            rel_fname = self.get_rel_fname(fname)
+            lines.append(f"Added {rel_fname} to the chat (read-only).")
+
         if self.done_messages:
             lines.append("Restored previous conversation history.")
 
+        if self.io.multiline_mode:
+            lines.append("Multiline mode: Enabled. Enter inserts newline, Alt-Enter submits text")
+
         return lines
 
     def __init__(
@@ -345,7 +359,6 @@ class Coder:
             self.done_messages = []
 
         self.io = io
-        self.stream = stream
 
         self.shell_commands = []
 
@@ -360,6 +373,8 @@ class Coder:
 
         self.main_model = main_model
 
+        self.stream = stream and main_model.streaming
+
         if cache_prompts and self.main_model.cache_control:
             self.add_cache_headers = True
 
@@ -450,6 +465,7 @@ class Coder:
 
         self.summarizer_thread = None
         self.summarized_done_messages = []
+        self.summarizing_messages = None
 
         if not self.done_messages and restore_chat_history:
             history_md = self.io.read_text(self.io.chat_history_file)
@@ -516,7 +532,7 @@ class Coder:
             return False
 
         # only show pretty output if fences are the normal triple-backtick
-        if self.fence != self.fences[0]:
+        if self.fence[0][0] != "`":
             return False
 
         return True
@@ -610,9 +626,19 @@ class Coder:
     def get_ident_filename_matches(self, idents):
         all_fnames = defaultdict(set)
         for fname in self.get_all_relative_files():
-            base = Path(fname).with_suffix("").name.lower()
-            if len(base) >= 5:
-                all_fnames[base].add(fname)
+            # Skip empty paths or just '.'
+            if not fname or fname == ".":
+                continue
+
+            try:
+                # Handle dotfiles properly
+                path = Path(fname)
+                base = path.stem.lower()  # Use stem instead of with_suffix("").name
+                if len(base) >= 5:
+                    all_fnames[base].add(fname)
+            except ValueError:
+                # Skip paths that can't be processed
+                continue
 
         matches = set()
         for ident in idents:
@@ -923,8 +949,9 @@ class Coder:
         self.summarizer_thread.start()
 
     def summarize_worker(self):
+        self.summarizing_messages = list(self.done_messages)
         try:
-            self.summarized_done_messages = self.summarizer.summarize(self.done_messages)
+            self.summarized_done_messages = self.summarizer.summarize(self.summarizing_messages)
         except ValueError as err:
             self.io.tool_warning(err.args[0])
 
@@ -938,7 +965,9 @@ class Coder:
         self.summarizer_thread.join()
         self.summarizer_thread = None
 
-        self.done_messages = self.summarized_done_messages
+        if self.summarizing_messages == self.done_messages:
+            self.done_messages = self.summarized_done_messages
+        self.summarizing_messages = None
         self.summarized_done_messages = []
 
     def move_back_cur_messages(self, message):
@@ -1032,8 +1061,16 @@ class Coder:
         else:
             language = "the same language they are using"
 
+        if self.fence[0] == "`" * 4:
+            quad_backtick_reminder = (
+                "\nIMPORTANT: Use *quadruple* backticks ```` as fences, not triple backticks!\n"
+            )
+        else:
+            quad_backtick_reminder = ""
+
         prompt = prompt.format(
             fence=self.fence,
+            quad_backtick_reminder=quad_backtick_reminder,
             lazy_prompt=lazy_prompt,
             platform=platform_text,
             shell_cmd_prompt=shell_cmd_prompt,
@@ -1207,6 +1244,29 @@ class Coder:
 
         return chunks
 
+    def check_tokens(self, messages):
+        """Check if the messages will fit within the model's token limits."""
+        input_tokens = self.main_model.token_count(messages)
+        max_input_tokens = self.main_model.info.get("max_input_tokens") or 0
+
+        if max_input_tokens and input_tokens >= max_input_tokens:
+            self.io.tool_error(
+                f"Your estimated chat context of {input_tokens:,} tokens exceeds the"
+                f" {max_input_tokens:,} token limit for {self.main_model.name}!"
+            )
+            self.io.tool_output("To reduce the chat context:")
+            self.io.tool_output("- Use /drop to remove unneeded files from the chat")
+            self.io.tool_output("- Use /clear to clear the chat history")
+            self.io.tool_output("- Break your code into smaller files")
+            self.io.tool_output(
+                "It's probably safe to try and send the request, most providers won't charge if"
+                " the context limit is exceeded."
+            )
+
+            if not self.io.confirm_ask("Try to proceed anyway?"):
+                return False
+        return True
+
     def send_message(self, inp):
         self.event("message_send_starting")
 
@@ -1216,6 +1276,8 @@ class Coder:
 
         chunks = self.format_messages()
         messages = chunks.all_messages()
+        if not self.check_tokens(messages):
+            return
         self.warm_cache(chunks)
 
         if self.verbose:
@@ -1276,7 +1338,7 @@ class Coder:
                         exhausted = True
                         break
 
-                    self.multi_response_content = self.get_multi_response_content()
+                    self.multi_response_content = self.get_multi_response_content_in_progress()
 
                     if messages[-1]["role"] == "assistant":
                         messages[-1]["content"] = self.multi_response_content
@@ -1296,14 +1358,27 @@ class Coder:
                 self.live_incremental_response(True)
                 self.mdstream = None
 
-            self.partial_response_content = self.get_multi_response_content(True)
+            self.partial_response_content = self.get_multi_response_content_in_progress(True)
+            self.partial_response_content = self.main_model.remove_reasoning_content(
+                self.partial_response_content
+            )
             self.multi_response_content = ""
 
         self.io.tool_output()
 
         self.show_usage_report()
 
+        self.add_assistant_reply_to_cur_messages()
+
         if exhausted:
+            if self.cur_messages and self.cur_messages[-1]["role"] == "user":
+                self.cur_messages += [
+                    dict(
+                        role="assistant",
+                        content="FinishReasonLength exception: you sent too many tokens",
+                    ),
+                ]
+
             self.show_exhausted_error()
             self.num_exhausted_context_windows += 1
             return
@@ -1334,14 +1409,17 @@ class Coder:
                 interrupted = True
 
         if interrupted:
-            content += "\n^C KeyboardInterrupt"
-            self.cur_messages += [dict(role="assistant", content=content)]
+            if self.cur_messages and self.cur_messages[-1]["role"] == "user":
+                self.cur_messages[-1]["content"] += "\n^C KeyboardInterrupt"
+            else:
+                self.cur_messages += [dict(role="user", content="^C KeyboardInterrupt")]
+            self.cur_messages += [
+                dict(role="assistant", content="I see that you interrupted my previous reply.")
+            ]
             return
 
         edited = self.apply_updates()
 
-        self.update_cur_messages()
-
         if edited:
             self.aider_edited_files.update(edited)
             saved_message = self.auto_commit(edited)
@@ -1362,7 +1440,6 @@ class Coder:
                 ok = self.io.confirm_ask("Attempt to fix lint errors?")
                 if ok:
                     self.reflected_message = lint_errors
-                    self.update_cur_messages()
                     return
 
         shared_output = self.run_shell_commands()
@@ -1379,7 +1456,6 @@ class Coder:
                 ok = self.io.confirm_ask("Attempt to fix test errors?")
                 if ok:
                     self.reflected_message = test_errors
-                    self.update_cur_messages()
                     return
 
     def reply_completed(self):
@@ -1455,7 +1531,7 @@ class Coder:
 
         return res
 
-    def update_cur_messages(self):
+    def add_assistant_reply_to_cur_messages(self):
         if self.partial_response_content:
             self.cur_messages += [dict(role="assistant", content=self.partial_response_content)]
         if self.partial_response_function_call:
@@ -1471,7 +1547,7 @@ class Coder:
         words = set(word for word in content.split())
 
         # drop sentence punctuation from the end
-        words = set(word.rstrip(",.!;:") for word in words)
+        words = set(word.rstrip(",.!;:?") for word in words)
 
         # strip away all kinds of quotes
         quotes = "".join(['"', "'", "`"])
@@ -1521,7 +1597,9 @@ class Coder:
         added_fnames = []
         group = ConfirmGroup(new_mentions)
         for rel_fname in sorted(new_mentions):
-            if self.io.confirm_ask(f"Add {rel_fname} to the chat?", group=group, allow_never=True):
+            if self.io.confirm_ask(
+                "Add file to the chat?", subject=rel_fname, group=group, allow_never=True
+            ):
                 self.add_rel_fname(rel_fname)
                 added_fnames.append(rel_fname)
             else:
@@ -1539,20 +1617,13 @@ class Coder:
 
         self.io.log_llm_history("TO LLM", format_messages(messages))
 
-        if self.main_model.use_temperature:
-            temp = self.temperature
-        else:
-            temp = None
-
         completion = None
         try:
-            hash_object, completion = send_completion(
-                model.name,
+            hash_object, completion = model.send_completion(
                 messages,
                 functions,
                 self.stream,
-                temp,
-                extra_params=model.extra_params,
+                self.temperature,
             )
             self.chat_completion_call_hashes.append(hash_object.hexdigest())
 
@@ -1679,7 +1750,7 @@ class Coder:
         self.mdstream.update(show_resp, final=final)
 
     def render_incremental_response(self, final):
-        return self.get_multi_response_content()
+        return self.get_multi_response_content_in_progress()
 
     def calculate_and_show_tokens_and_cost(self, messages, completion=None):
         prompt_tokens = 0
@@ -1802,12 +1873,13 @@ class Coder:
         self.message_tokens_sent = 0
         self.message_tokens_received = 0
 
-    def get_multi_response_content(self, final=False):
+    def get_multi_response_content_in_progress(self, final=False):
         cur = self.multi_response_content or ""
         new = self.partial_response_content or ""
 
         if new.rstrip() != new and not final:
             new = new.rstrip()
+
         return cur + new
 
     def get_rel_fname(self, fname):

aider/coders/editblock_coder.py

@@ -401,6 +401,9 @@ missing_filename_err = (
     " {fence[0]}"
 )
 
+# Always be willing to treat triple-backticks as a fence when searching for filenames
+triple_backticks = "`" * 3
+
 
 def strip_filename(filename, fence):
     filename = filename.strip()
@@ -409,7 +412,7 @@ def strip_filename(filename, fence):
         return
 
     start_fence = fence[0]
-    if filename.startswith(start_fence):
+    if filename.startswith(start_fence) or filename.startswith(triple_backticks):
         return
 
     filename = filename.rstrip(":")
@@ -546,7 +549,7 @@ def find_filename(lines, fence, valid_fnames):
             filenames.append(filename)
 
         # Only continue as long as we keep seeing fences
-        if not line.startswith(fence[0]):
+        if not line.startswith(fence[0]) and not line.startswith(triple_backticks):
             break
 
     if not filenames:

aider/coders/editblock_prompts.py

@@ -157,7 +157,7 @@ Every *SEARCH/REPLACE block* must use this format:
 8. The closing fence: {fence[1]}
 
 Use the *FULL* file path, as shown to you by the user.
-
+{quad_backtick_reminder}
 Every *SEARCH* section must *EXACTLY MATCH* the existing file content, character for character, including all comments, docstrings, etc.
 If the file contains code or other data wrapped/escaped in json/xml/quotes or other containers, you need to propose edits to the literal contents of the file, including the container markup.
 
@@ -183,6 +183,9 @@ If you want to put code in a new file, use a *SEARCH/REPLACE block* with:
 
 To rename files which have been added to the chat, use shell commands at the end of your response.
 
+If the user just says something like "ok" or "go ahead" or "do that" they probably want you to make SEARCH/REPLACE blocks for the code changes you just proposed.
+The user will say when they've applied your edits. If they haven't explicitly confirmed the edits have been applied, they probably want proper SEARCH/REPLACE blocks.
+
 {lazy_prompt}
 ONLY EVER RETURN CODE IN A *SEARCH/REPLACE BLOCK*!
 {shell_cmd_reminder}

aider/coders/search_replace.py

@@ -3,7 +3,11 @@
 import sys
 from pathlib import Path
 
-import git
+try:
+    import git
+except ImportError:
+    git = None
+
 from diff_match_patch import diff_match_patch
 from tqdm import tqdm
 

aider/coders/single_wholefile_func_coder.py

@@ -38,7 +38,7 @@ class SingleWholeFileFunctionCoder(Coder):
         self.gpt_prompts = SingleWholeFileFunctionPrompts()
         super().__init__(*args, **kwargs)
 
-    def update_cur_messages(self, edited):
+    def add_assistant_reply_to_cur_messages(self, edited):
         if edited:
             self.cur_messages += [
                 dict(role="assistant", content=self.gpt_prompts.redacted_edit_message)

aider/coders/wholefile_coder.py

@@ -17,10 +17,10 @@ class WholeFileCoder(Coder):
         try:
             return self.get_edits(mode="diff")
         except ValueError:
-            return self.get_multi_response_content()
+            return self.get_multi_response_content_in_progress()
 
     def get_edits(self, mode="update"):
-        content = self.get_multi_response_content()
+        content = self.get_multi_response_content_in_progress()
 
         chat_files = self.get_inchat_relative_files()
 

aider/coders/wholefile_func_coder.py

@@ -49,7 +49,7 @@ class WholeFileFunctionCoder(Coder):
         self.gpt_prompts = WholeFileFunctionPrompts()
         super().__init__(*args, **kwargs)
 
-    def update_cur_messages(self, edited):
+    def add_assistant_reply_to_cur_messages(self, edited):
         if edited:
             self.cur_messages += [
                 dict(role="assistant", content=self.gpt_prompts.redacted_edit_message)

aider/commands.py

@@ -52,6 +52,8 @@ class Commands:
         io,
         coder,
         voice_language=None,
+        voice_input_device=None,
+        voice_format=None,
         verify_ssl=True,
         args=None,
         parser=None,
@@ -69,6 +71,8 @@ class Commands:
             voice_language = None
 
         self.voice_language = voice_language
+        self.voice_format = voice_format
+        self.voice_input_device = voice_input_device
 
         self.help = None
         self.editor = editor
@@ -77,7 +81,7 @@ class Commands:
         "Switch to a new LLM"
 
         model_name = args.strip()
-        model = models.Model(model_name)
+        model = models.Model(model_name, weak_model=self.coder.main_model.weak_model.name)
         models.sanity_check_models(self.io, model)
         raise SwitchCoder(main_model=model)
 
@@ -752,6 +756,7 @@ class Commands:
 
             if self.io.confirm_ask(f"No files matched '{word}'. Do you want to create {fname}?"):
                 try:
+                    fname.parent.mkdir(parents=True, exist_ok=True)
                     fname.touch()
                     all_matched_files.add(str(fname))
                 except OSError as e:
@@ -914,10 +919,14 @@ class Commands:
         if combined_output is None:
             return
 
+        # Calculate token count of output
+        token_count = self.coder.main_model.token_count(combined_output)
+        k_tokens = token_count / 1000
+
         if add_on_nonzero_exit:
             add = exit_status != 0
         else:
-            add = self.io.confirm_ask("Add command output to the chat?")
+            add = self.io.confirm_ask(f"Add {k_tokens:.1f}k tokens of command output to the chat?")
 
         if add:
             num_lines = len(combined_output.strip().splitlines())
@@ -935,7 +944,7 @@ class Commands:
             ]
 
             if add and exit_status != 0:
-                self.io.placeholder = "Fix that"
+                self.io.placeholder = "What's wrong? Fix"
 
     def cmd_exit(self, args):
         "Exit the application"
@@ -1009,7 +1018,7 @@ class Commands:
             return
 
         self.coder.event("interactive help")
-        from aider.coders import Coder
+        from aider.coders.base_coder import Coder
 
         if not self.help:
             res = install_help_extra(self.io)
@@ -1053,23 +1062,23 @@ class Commands:
         )
 
     def cmd_ask(self, args):
-        "Ask questions about the code base without editing any files"
+        """Ask questions about the code base without editing any files. If no prompt provided, switches to ask mode."""  # noqa
         return self._generic_chat_command(args, "ask")
 
     def cmd_code(self, args):
-        "Ask for changes to your code"
+        """Ask for changes to your code. If no prompt provided, switches to code mode."""  # noqa
         return self._generic_chat_command(args, self.coder.main_model.edit_format)
 
     def cmd_architect(self, args):
-        "Enter architect mode to discuss high-level design and architecture"
+        """Enter architect/editor mode using 2 different models. If no prompt provided, switches to architect/editor mode."""  # noqa
         return self._generic_chat_command(args, "architect")
 
     def _generic_chat_command(self, args, edit_format):
         if not args.strip():
-            self.io.tool_error(f"Please provide a question or topic for the {edit_format} chat.")
-            return
+            # Switch to the corresponding chat mode if no args provided
+            return self.cmd_chat_mode(edit_format)
 
-        from aider.coders import Coder
+        from aider.coders.base_coder import Coder
 
         coder = Coder.create(
             io=self.io,
@@ -1117,7 +1126,7 @@ class Commands:
                 return
             try:
                 self.voice = voice.Voice(
-                    audio_format=self.args.voice_format, device_name=self.args.voice_input_device
+                    audio_format=self.voice_format or "wav", device_name=self.voice_input_device
                 )
             except voice.SoundDeviceError:
                 self.io.tool_error(
@@ -1125,36 +1134,14 @@ class Commands:
                 )
                 return
 
-        history_iter = self.io.get_input_history()
-
-        history = []
-        size = 0
-        for line in history_iter:
-            if line.startswith("/"):
-                continue
-            if line in history:
-                continue
-            if size + len(line) > 1024:
-                break
-            size += len(line)
-            history.append(line)
-
-        history.reverse()
-        history = "\n".join(history)
-
         try:
-            text = self.voice.record_and_transcribe(history, language=self.voice_language)
+            text = self.voice.record_and_transcribe(None, language=self.voice_language)
         except litellm.OpenAIError as err:
             self.io.tool_error(f"Unable to use OpenAI whisper model: {err}")
             return
 
         if text:
-            self.io.add_to_input_history(text)
-            self.io.print()
-            self.io.user_input(text, log_only=False)
-            self.io.print()
-
-        return text
+            self.io.placeholder = text
 
     def cmd_paste(self, args):
         """Paste image/text from the clipboard into the chat.\
@@ -1331,7 +1318,12 @@ class Commands:
                 continue
 
             self.io.tool_output(f"\nExecuting: {cmd}")
-            self.run(cmd)
+            try:
+                self.run(cmd)
+            except SwitchCoder:
+                self.io.tool_error(
+                    f"Command '{cmd}' is only supported in interactive mode, skipping."
+                )
 
     def completions_raw_save(self, document, complete_event):
         return self.completions_raw_read_only(document, complete_event)
@@ -1363,6 +1355,10 @@ class Commands:
         except Exception as e:
             self.io.tool_error(f"Error saving commands to file: {e}")
 
+    def cmd_multiline_mode(self, args):
+        "Toggle multiline mode (swaps behavior of Enter and Meta+Enter)"
+        self.io.toggle_multiline_mode()
+
     def cmd_copy(self, args):
         "Copy the last assistant message to the clipboard"
         all_messages = self.coder.done_messages + self.coder.cur_messages
@@ -1435,16 +1431,25 @@ class Commands:
                 else:
                     markdown += content + "\n\n"
 
-        markdown += """
+        args = args or ""
+        markdown += f"""
 Just tell me how to edit the files to make the changes.
 Don't give me back entire files.
 Just show me the edits I need to make.
 
-
+{args}
 """
 
-        pyperclip.copy(markdown)
-        self.io.tool_output("Copied code context to clipboard.")
+        try:
+            pyperclip.copy(markdown)
+            self.io.tool_output("Copied code context to clipboard.")
+        except pyperclip.PyperclipException as e:
+            self.io.tool_error(f"Failed to copy to clipboard: {str(e)}")
+            self.io.tool_output(
+                "You may need to install xclip or xsel on Linux, or pbcopy on macOS."
+            )
+        except Exception as e:
+            self.io.tool_error(f"An unexpected error occurred while copying to clipboard: {str(e)}")
 
 
 def expand_subdir(file_path):

aider/exceptions.py

@@ -1,5 +1,7 @@
 from dataclasses import dataclass
 
+from aider.dump import dump  # noqa: F401
+
 
 @dataclass
 class ExInfo:
@@ -50,6 +52,7 @@ EXCEPTIONS = [
 
 class LiteLLMExceptions:
     exceptions = dict()
+    exception_info = {exi.name: exi for exi in EXCEPTIONS}
 
     def __init__(self):
         self._load()
@@ -58,20 +61,13 @@ class LiteLLMExceptions:
         import litellm
 
         for var in dir(litellm):
-            if not var.endswith("Error"):
-                continue
-
-            ex_info = None
-            for exi in EXCEPTIONS:
-                if var == exi.name:
-                    ex_info = exi
-                    break
-
-            if strict and not ex_info:
-                raise ValueError(f"{var} is in litellm but not in aider's exceptions list")
+            if var.endswith("Error"):
+                if var not in self.exception_info:
+                    raise ValueError(f"{var} is in litellm but not in aider's exceptions list")
 
+        for var in self.exception_info:
             ex = getattr(litellm, var)
-            self.exceptions[ex] = ex_info
+            self.exceptions[ex] = self.exception_info[var]
 
     def exceptions_tuple(self):
         return tuple(self.exceptions)

aider/help_pats.py

@@ -1,4 +1,7 @@
+# This needs to sync with MANIFEST.in
+
 exclude_website_pats = [
+    "**/.DS_Store",
     "examples/**",
     "_posts/**",
     "HISTORY.md",
@@ -7,5 +10,4 @@ exclude_website_pats = [
     "docs/unified-diffs.md",
     "docs/leaderboards/index.md",
     "assets/**",
-    "**/.DS_Store",
 ]

aider/history.py

@@ -2,7 +2,6 @@ import argparse
 
 from aider import models, prompts
 from aider.dump import dump  # noqa: F401
-from aider.sendchat import simple_send_with_retries
 
 
 class ChatSummary:
@@ -26,6 +25,12 @@ class ChatSummary:
         return sized
 
     def summarize(self, messages, depth=0):
+        messages = self.summarize_real(messages)
+        if messages and messages[-1]["role"] != "assistant":
+            messages.append(dict(role="assistant", content="Ok."))
+        return messages
+
+    def summarize_real(self, messages, depth=0):
         if not self.models:
             raise ValueError("No models available for summarization")
 
@@ -88,7 +93,7 @@ class ChatSummary:
         if summary_tokens + tail_tokens < self.max_tokens:
             return result
 
-        return self.summarize(result, depth + 1)
+        return self.summarize_real(result, depth + 1)
 
     def summarize_all(self, messages):
         content = ""
@@ -108,7 +113,7 @@ class ChatSummary:
 
         for model in self.models:
             try:
-                summary = simple_send_with_retries(model, summarize_messages)
+                summary = model.simple_send_with_retries(summarize_messages)
                 if summary is not None:
                     summary = prompts.summary_prefix + summary
                     return [dict(role="user", content=summary)]

aider/io.py

@@ -1,4 +1,5 @@
 import base64
+import functools
 import os
 import signal
 import time
@@ -12,11 +13,12 @@ from pathlib import Path
 from prompt_toolkit.completion import Completer, Completion, ThreadedCompleter
 from prompt_toolkit.cursor_shapes import ModalCursorShapeConfig
 from prompt_toolkit.enums import EditingMode
-from prompt_toolkit.filters import Condition
+from prompt_toolkit.filters import Condition, is_searching
 from prompt_toolkit.history import FileHistory
 from prompt_toolkit.key_binding import KeyBindings
 from prompt_toolkit.keys import Keys
 from prompt_toolkit.lexers import PygmentsLexer
+from prompt_toolkit.output.vt100 import is_dumb_terminal
 from prompt_toolkit.shortcuts import CompleteStyle, PromptSession
 from prompt_toolkit.styles import Style
 from pygments.lexers import MarkdownLexer, guess_lexer_for_filename
@@ -33,6 +35,23 @@ from .dump import dump  # noqa: F401
 from .utils import is_image_file
 
 
+def restore_multiline(func):
+    """Decorator to restore multiline mode after function execution"""
+
+    @functools.wraps(func)
+    def wrapper(self, *args, **kwargs):
+        orig_multiline = self.multiline_mode
+        self.multiline_mode = False
+        try:
+            return func(self, *args, **kwargs)
+        except Exception:
+            raise
+        finally:
+            self.multiline_mode = orig_multiline
+
+    return wrapper
+
+
 @dataclass
 class ConfirmGroup:
     preference: str = None
@@ -197,16 +216,20 @@ class InputOutput:
         completion_menu_current_bg_color=None,
         code_theme="default",
         encoding="utf-8",
+        line_endings="platform",
         dry_run=False,
         llm_history_file=None,
         editingmode=EditingMode.EMACS,
         fancy_input=True,
         file_watcher=None,
+        multiline_mode=False,
+        root=".",
     ):
         self.placeholder = None
         self.interrupted = False
         self.never_prompts = set()
         self.editingmode = editingmode
+        self.multiline_mode = multiline_mode
         no_color = os.environ.get("NO_COLOR")
         if no_color is not None and no_color != "":
             pretty = False
@@ -240,14 +263,29 @@ class InputOutput:
             self.chat_history_file = None
 
         self.encoding = encoding
+        valid_line_endings = {"platform", "lf", "crlf"}
+        if line_endings not in valid_line_endings:
+            raise ValueError(
+                f"Invalid line_endings value: {line_endings}. "
+                f"Must be one of: {', '.join(valid_line_endings)}"
+            )
+        self.newline = (
+            None if line_endings == "platform" else "\n" if line_endings == "lf" else "\r\n"
+        )
         self.dry_run = dry_run
 
         current_time = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
         self.append_chat_history(f"\n# aider chat started at {current_time}\n\n")
 
         self.prompt_session = None
+        self.is_dumb_terminal = is_dumb_terminal()
+
+        if self.is_dumb_terminal:
+            self.pretty = False
+            fancy_input = False
+
         if fancy_input:
-            # Initialize PromptSession
+            # Initialize PromptSession only if we have a capable terminal
             session_kwargs = {
                 "input": self.input,
                 "output": self.output,
@@ -266,8 +304,11 @@ class InputOutput:
                 self.tool_error(f"Can't initialize prompt toolkit: {err}")  # non-pretty
         else:
             self.console = Console(force_terminal=False, no_color=True)  # non-pretty
+            if self.is_dumb_terminal:
+                self.tool_output("Detected dumb terminal, disabling fancy input and pretty output.")
 
         self.file_watcher = file_watcher
+        self.root = root
 
     def _get_style(self):
         style_dict = {}
@@ -329,10 +370,6 @@ class InputOutput:
         try:
             with open(str(filename), "r", encoding=self.encoding) as f:
                 return f.read()
-        except OSError as err:
-            if not silent:
-                self.tool_error(f"{filename}: unable to read: {err}")
-            return
         except FileNotFoundError:
             if not silent:
                 self.tool_error(f"{filename}: file not found error")
@@ -341,6 +378,10 @@ class InputOutput:
             if not silent:
                 self.tool_error(f"{filename}: is a directory")
             return
+        except OSError as err:
+            if not silent:
+                self.tool_error(f"{filename}: unable to read: {err}")
+            return
         except UnicodeError as e:
             if not silent:
                 self.tool_error(f"{filename}: {e}")
@@ -362,7 +403,7 @@ class InputOutput:
         delay = initial_delay
         for attempt in range(max_retries):
             try:
-                with open(str(filename), "w", encoding=self.encoding) as f:
+                with open(str(filename), "w", encoding=self.encoding, newline=self.newline) as f:
                     f.write(content)
                 return  # Successfully wrote the file
             except PermissionError as err:
@@ -412,6 +453,8 @@ class InputOutput:
             show = self.format_files_for_input(rel_fnames, rel_read_only_fnames)
         if edit_format:
             show += edit_format
+        if self.multiline_mode:
+            show += (" " if edit_format else "") + "multi"
         show += "> "
 
         inp = ""
@@ -456,9 +499,25 @@ class InputOutput:
             "Navigate forward through history"
             event.current_buffer.history_forward()
 
-        @kb.add("escape", "c-m", eager=True)
+        @kb.add("enter", eager=True, filter=~is_searching)
         def _(event):
-            event.current_buffer.insert_text("\n")
+            "Handle Enter key press"
+            if self.multiline_mode:
+                # In multiline mode, Enter adds a newline
+                event.current_buffer.insert_text("\n. ")
+            else:
+                # In normal mode, Enter submits
+                event.current_buffer.validate_and_handle()
+
+        @kb.add("escape", "enter", eager=True, filter=~is_searching)  # This is Alt+Enter
+        def _(event):
+            "Handle Alt+Enter key press"
+            if self.multiline_mode:
+                # In multiline mode, Alt+Enter submits
+                event.current_buffer.validate_and_handle()
+            else:
+                # In normal mode, Alt+Enter adds a newline
+                event.current_buffer.insert_text("\n. ")
 
         while True:
             if multiline_input:
@@ -485,6 +544,7 @@ class InputOutput:
                         complete_style=CompleteStyle.MULTI_COLUMN,
                         style=style,
                         key_bindings=kb,
+                        complete_while_typing=True,
                     )
                 else:
                     line = input(show)
@@ -620,6 +680,7 @@ class InputOutput:
             return True
         return False
 
+    @restore_multiline
     def confirm_ask(
         self,
         question,
@@ -641,19 +702,22 @@ class InputOutput:
         if group:
             allow_never = True
 
-        valid_responses = ["yes", "no"]
+        valid_responses = ["yes", "no", "skip", "all"]
         options = " (Y)es/(N)o"
         if group:
             if not explicit_yes_required:
                 options += "/(A)ll"
-                valid_responses.append("all")
             options += "/(S)kip all"
-            valid_responses.append("skip")
         if allow_never:
             options += "/(D)on't ask again"
             valid_responses.append("don't")
 
-        question += options + " [Yes]: "
+        if default.lower().startswith("y"):
+            question += options + " [Yes]: "
+        elif default.lower().startswith("n"):
+            question += options + " [No]: "
+        else:
+            question += options + f" [{default}]: "
 
         if subject:
             self.tool_output()
@@ -686,12 +750,13 @@ class InputOutput:
                     res = self.prompt_session.prompt(
                         question,
                         style=style,
+                        complete_while_typing=False,
                     )
                 else:
                     res = input(question)
 
                 if not res:
-                    res = "y"  # Default to Yes if no input
+                    res = default
                     break
                 res = res.lower()
                 good = any(valid_response.startswith(res) for valid_response in valid_responses)
@@ -728,6 +793,7 @@ class InputOutput:
 
         return is_yes
 
+    @restore_multiline
     def prompt_ask(self, question, default="", subject=None):
         self.num_user_asks += 1
 
@@ -743,7 +809,12 @@ class InputOutput:
             res = "no"
         else:
             if self.prompt_session:
-                res = self.prompt_session.prompt(question + " ", default=default, style=style)
+                res = self.prompt_session.prompt(
+                    question + " ",
+                    default=default,
+                    style=style,
+                    complete_while_typing=True,
+                )
             else:
                 res = input(question + " ")
 
@@ -763,9 +834,17 @@ class InputOutput:
                 hist = message.strip() if strip else message
                 self.append_chat_history(hist, linebreak=True, blockquote=True)
 
-        message = Text(message)
+        if not isinstance(message, Text):
+            message = Text(message)
         style = dict(style=color) if self.pretty and color else dict()
-        self.console.print(message, **style)
+        try:
+            self.console.print(message, **style)
+        except UnicodeEncodeError:
+            # Fallback to ASCII-safe output
+            if isinstance(message, Text):
+                message = message.plain
+            message = str(message).encode("ascii", errors="replace").decode("ascii")
+            self.console.print(message, **style)
 
     def tool_error(self, message="", strip=True):
         self.num_error_outputs += 1
@@ -821,6 +900,18 @@ class InputOutput:
     def print(self, message=""):
         print(message)
 
+    def toggle_multiline_mode(self):
+        """Toggle between normal and multiline input modes"""
+        self.multiline_mode = not self.multiline_mode
+        if self.multiline_mode:
+            self.tool_output(
+                "Multiline mode: Enabled. Enter inserts newline, Alt-Enter submits text"
+            )
+        else:
+            self.tool_output(
+                "Multiline mode: Disabled. Alt-Enter inserts newline, Enter submits text"
+            )
+
     def append_chat_history(self, text, linebreak=False, blockquote=False, strip=True):
         if blockquote:
             if strip:
@@ -862,7 +953,13 @@ class InputOutput:
         editable_files = [f for f in sorted(rel_fnames) if f not in rel_read_only_fnames]
 
         if read_only_files:
-            files_with_label = ["Readonly:"] + read_only_files
+            # Use shorter of abs/rel paths for readonly files
+            ro_paths = []
+            for rel_path in read_only_files:
+                abs_path = os.path.abspath(os.path.join(self.root, rel_path))
+                ro_paths.append(abs_path if len(abs_path) < len(rel_path) else rel_path)
+
+            files_with_label = ["Readonly:"] + ro_paths
             read_only_output = StringIO()
             Console(file=read_only_output, force_terminal=False).print(Columns(files_with_label))
             read_only_lines = read_only_output.getvalue().splitlines()

aider/linter.py

@@ -11,6 +11,7 @@ from grep_ast import TreeContext, filename_to_lang
 from tree_sitter_languages import get_parser  # noqa: E402
 
 from aider.dump import dump  # noqa: F401
+from aider.run_cmd import run_cmd_subprocess  # noqa: F401
 
 # tree_sitter is throwing a FutureWarning
 warnings.simplefilter("ignore", category=FutureWarning)
@@ -44,26 +45,22 @@ class Linter:
 
     def run_cmd(self, cmd, rel_fname, code):
         cmd += " " + rel_fname
-        cmd = cmd.split()
 
+        returncode = 0
+        stdout = ""
         try:
-            process = subprocess.Popen(
+            returncode, stdout = run_cmd_subprocess(
                 cmd,
-                stdout=subprocess.PIPE,
-                stderr=subprocess.STDOUT,
-                encoding=self.encoding,
-                errors="replace",
                 cwd=self.root,
+                encoding=self.encoding,
             )
         except OSError as err:
             print(f"Unable to execute lint command: {err}")
             return
-        stdout, _ = process.communicate()
         errors = stdout
-        if process.returncode == 0:
+        if returncode == 0:
             return  # zero exit status
 
-        cmd = " ".join(cmd)
         res = f"## Running: {cmd}\n\n"
         res += errors
 

aider/llm.py

@@ -2,6 +2,8 @@ import importlib
 import os
 import warnings
 
+from aider.dump import dump  # noqa: F401
+
 warnings.filterwarnings("ignore", category=UserWarning, module="pydantic")
 
 AIDER_SITE_URL = "https://aider.chat"

aider/main.py

@@ -9,7 +9,11 @@ import webbrowser
 from dataclasses import fields
 from pathlib import Path
 
-import git
+try:
+    import git
+except ImportError:
+    git = None
+
 import importlib_resources
 from dotenv import load_dotenv
 from prompt_toolkit.enums import EditingMode
@@ -93,6 +97,9 @@ def make_new_repo(git_root, io):
 
 
 def setup_git(git_root, io):
+    if git is None:
+        return
+
     try:
         cwd = Path.cwd()
     except OSError:
@@ -106,7 +113,9 @@ def setup_git(git_root, io):
         except ANY_GIT_ERROR:
             pass
     elif cwd == Path.home():
-        io.tool_warning("You should probably run aider in a directory, not your home dir.")
+        io.tool_warning(
+            "You should probably run aider in your project's directory, not your home dir."
+        )
         return
     elif cwd and io.confirm_ask(
         "No git repo found, create one to track aider's changes (recommended)?"
@@ -149,39 +158,39 @@ def check_gitignore(git_root, io, ask=True):
 
     try:
         repo = git.Repo(git_root)
-        if repo.ignored(".aider") and repo.ignored(".env"):
+        patterns_to_add = []
+
+        if not repo.ignored(".aider"):
+            patterns_to_add.append(".aider*")
+
+        env_path = Path(git_root) / ".env"
+        if env_path.exists() and not repo.ignored(".env"):
+            patterns_to_add.append(".env")
+
+        if not patterns_to_add:
             return
-    except ANY_GIT_ERROR:
-        pass
 
-    patterns = [".aider*", ".env"]
-    patterns_to_add = []
-
-    gitignore_file = Path(git_root) / ".gitignore"
-    if gitignore_file.exists():
-        try:
-            content = io.read_text(gitignore_file)
-            if content is None:
+        gitignore_file = Path(git_root) / ".gitignore"
+        if gitignore_file.exists():
+            try:
+                content = io.read_text(gitignore_file)
+                if content is None:
+                    return
+                if not content.endswith("\n"):
+                    content += "\n"
+            except OSError as e:
+                io.tool_error(f"Error when trying to read {gitignore_file}: {e}")
                 return
-            existing_lines = content.splitlines()
-            for pat in patterns:
-                if pat not in existing_lines:
-                    patterns_to_add.append(pat)
-        except OSError as e:
-            io.tool_error(f"Error when trying to read {gitignore_file}: {e}")
+        else:
+            content = ""
+    except ANY_GIT_ERROR:
+        return
+
+    if ask:
+        io.tool_output("You can skip this check with --no-gitignore")
+        if not io.confirm_ask(f"Add {', '.join(patterns_to_add)} to .gitignore (recommended)?"):
             return
-    else:
-        content = ""
-        patterns_to_add = patterns
 
-    if not patterns_to_add:
-        return
-
-    if ask and not io.confirm_ask(f"Add {', '.join(patterns_to_add)} to .gitignore (recommended)?"):
-        return
-
-    if content and not content.endswith("\n"):
-        content += "\n"
     content += "\n".join(patterns_to_add) + "\n"
 
     try:
@@ -205,6 +214,22 @@ def check_streamlit_install(io):
     )
 
 
+def write_streamlit_credentials():
+    from streamlit.file_util import get_streamlit_file_path
+
+    # See https://github.com/Aider-AI/aider/issues/772
+
+    credential_path = Path(get_streamlit_file_path()) / "credentials.toml"
+    if not os.path.exists(credential_path):
+        empty_creds = '[general]\nemail = ""\n'
+
+        os.makedirs(os.path.dirname(credential_path), exist_ok=True)
+        with open(credential_path, "w") as f:
+            f.write(empty_creds)
+    else:
+        print("Streamlit credentials already exist.")
+
+
 def launch_gui(args):
     from streamlit.web import cli
 
@@ -213,6 +238,9 @@ def launch_gui(args):
     print()
     print("CONTROL-C to exit...")
 
+    # Necessary so streamlit does not prompt the user for an email address.
+    write_streamlit_credentials()
+
     target = gui.__file__
 
     st_args = ["run", target]
@@ -350,18 +378,18 @@ def load_dotenv_files(git_root, dotenv_fname, encoding="utf-8"):
 
 
 def register_litellm_models(git_root, model_metadata_fname, io, verbose=False):
-    model_metatdata_files = []
+    model_metadata_files = []
 
     # Add the resource file path
     resource_metadata = importlib_resources.files("aider.resources").joinpath("model-metadata.json")
-    model_metatdata_files.append(str(resource_metadata))
+    model_metadata_files.append(str(resource_metadata))
 
-    model_metatdata_files += generate_search_path_list(
+    model_metadata_files += generate_search_path_list(
         ".aider.model.metadata.json", git_root, model_metadata_fname
     )
 
     try:
-        model_metadata_files_loaded = models.register_litellm_models(model_metatdata_files)
+        model_metadata_files_loaded = models.register_litellm_models(model_metadata_files)
         if len(model_metadata_files_loaded) > 0 and verbose:
             io.tool_output("Loaded model metadata from:")
             for model_metadata_file in model_metadata_files_loaded:
@@ -385,6 +413,12 @@ def sanity_check_repo(repo, io):
         if not repo.git_repo_error:
             return True
         error_msg = str(repo.git_repo_error)
+    except UnicodeDecodeError as exc:
+        error_msg = (
+            "Failed to read the Git repository. This issue is likely caused by a path encoded "
+            f'in a format different from the expected encoding "{sys.getfilesystemencoding()}".\n'
+            f"Internal error: {str(exc)}"
+        )
     except ANY_GIT_ERROR as exc:
         error_msg = str(exc)
         bad_ver = "version in (1, 2)" in error_msg
@@ -410,7 +444,9 @@ def main(argv=None, input=None, output=None, force_git_root=None, return_coder=F
     if argv is None:
         argv = sys.argv[1:]
 
-    if force_git_root:
+    if git is None:
+        git_root = None
+    elif force_git_root:
         git_root = force_git_root
     else:
         git_root = get_git_root()
@@ -457,6 +493,9 @@ def main(argv=None, input=None, output=None, force_git_root=None, return_coder=F
     # Parse again to include any arguments that might have been defined in .env
     args = parser.parse_args(argv)
 
+    if git is None:
+        args.git = False
+
     if args.analytics_disable:
         analytics = Analytics(permanently_disable=True)
         print("Analytics have been permanently disabled.")
@@ -470,8 +509,7 @@ def main(argv=None, input=None, output=None, force_git_root=None, return_coder=F
         litellm._lazy_module.aclient_session = httpx.AsyncClient(verify=False)
 
     if args.timeout:
-        litellm._load_litellm()
-        litellm._lazy_module.request_timeout = args.timeout
+        models.request_timeout = args.timeout
 
     if args.dark_mode:
         args.user_input_color = "#32FF32"
@@ -512,9 +550,11 @@ def main(argv=None, input=None, output=None, force_git_root=None, return_coder=F
             code_theme=args.code_theme,
             dry_run=args.dry_run,
             encoding=args.encoding,
+            line_endings=args.line_endings,
             llm_history_file=args.llm_history_file,
             editingmode=editing_mode,
             fancy_input=args.fancy_input,
+            multiline_mode=args.multiline,
         )
 
     io = get_io(args.pretty)
@@ -645,7 +685,7 @@ def main(argv=None, input=None, output=None, force_git_root=None, return_coder=F
     # We can't know the git repo for sure until after parsing the args.
     # If we guessed wrong, reparse because that changes things like
     # the location of the config.yml and history files.
-    if args.git and not force_git_root:
+    if args.git and not force_git_root and git is not None:
         right_repo_root = guessed_wrong_repo(io, git_root, fnames, git_dname)
         if right_repo_root:
             analytics.event("exit", reason="Recursing with correct repo")
@@ -707,9 +747,26 @@ def main(argv=None, input=None, output=None, force_git_root=None, return_coder=F
             models.MODEL_ALIASES[alias.strip()] = model.strip()
 
     if not args.model:
-        args.model = "gpt-4o-2024-08-06"
-        if os.environ.get("ANTHROPIC_API_KEY"):
-            args.model = "claude-3-5-sonnet-20241022"
+        # Select model based on available API keys
+        model_key_pairs = [
+            ("ANTHROPIC_API_KEY", "sonnet"),
+            ("DEEPSEEK_API_KEY", "deepseek"),
+            ("OPENROUTER_API_KEY", "openrouter/anthropic/claude-3.5-sonnet"),
+            ("OPENAI_API_KEY", "gpt-4o"),
+            ("GEMINI_API_KEY", "flash"),
+        ]
+
+        for env_key, model_name in model_key_pairs:
+            if os.environ.get(env_key):
+                args.model = model_name
+                io.tool_warning(
+                    f"Found {env_key} so using {model_name} since no --model was specified."
+                )
+                break
+        if not args.model:
+            io.tool_error("You need to specify a --model and an --api-key to use.")
+            io.offer_url(urls.models_and_keys, "Open documentation url for more info?")
+            return 1
 
     main_model = models.Model(
         args.model,
@@ -718,6 +775,14 @@ def main(argv=None, input=None, output=None, force_git_root=None, return_coder=F
         editor_edit_format=args.editor_edit_format,
     )
 
+    # add --reasoning-effort cli param
+    if args.reasoning_effort is not None:
+        if not getattr(main_model, "extra_params", None):
+            main_model.extra_params = {}
+        if "extra_body" not in main_model.extra_params:
+            main_model.extra_params["extra_body"] = {}
+        main_model.extra_params["extra_body"]["reasoning_effort"] = args.reasoning_effort
+
     if args.copy_paste and args.edit_format is None:
         if main_model.edit_format in ("diff", "whole"):
             main_model.edit_format = "editor-" + main_model.edit_format
@@ -782,6 +847,9 @@ def main(argv=None, input=None, output=None, force_git_root=None, return_coder=F
     commands = Commands(
         io,
         None,
+        voice_language=args.voice_language,
+        voice_input_device=args.voice_input_device,
+        voice_format=args.voice_format,
         verify_ssl=args.verify_ssl,
         args=args,
         parser=parser,
@@ -804,6 +872,11 @@ def main(argv=None, input=None, output=None, force_git_root=None, return_coder=F
             )
         args.stream = False
 
+    if args.map_tokens is None:
+        map_tokens = main_model.get_repo_map_tokens()
+    else:
+        map_tokens = args.map_tokens
+
     try:
         coder = Coder.create(
             main_model=main_model,
@@ -816,7 +889,7 @@ def main(argv=None, input=None, output=None, force_git_root=None, return_coder=F
             auto_commits=args.auto_commits,
             dirty_commits=args.dirty_commits,
             dry_run=args.dry_run,
-            map_tokens=args.map_tokens,
+            map_tokens=map_tokens,
             verbose=args.verbose,
             stream=args.stream,
             use_git=args.git,
@@ -859,7 +932,11 @@ def main(argv=None, input=None, output=None, force_git_root=None, return_coder=F
 
     if args.watch_files:
         file_watcher = FileWatcher(
-            coder, gitignores=ignores, verbose=args.verbose, analytics=analytics
+            coder,
+            gitignores=ignores,
+            verbose=args.verbose,
+            analytics=analytics,
+            root=str(Path.cwd()) if args.subtree_only else None,
         )
         coder.file_watcher = file_watcher
 
@@ -913,6 +990,9 @@ def main(argv=None, input=None, output=None, force_git_root=None, return_coder=F
             analytics.event("exit", reason="Failed to read apply content")
             return
         coder.partial_response_content = content
+        # For testing #2879
+        # from aider.coders.base_coder import all_fences
+        # coder.fence = all_fences[1]
         coder.apply_updates()
         analytics.event("exit", reason="Applied updates")
         return

aider/mdstream.py

@@ -10,10 +10,17 @@ from rich.text import Text
 
 from aider.dump import dump  # noqa: F401
 
-_text = """
+_text_prefix = """
 # Header
 
-Lorem Ipsum is simply dummy text of the printing and typesetting industry. Lorem Ipsum has been the industry's standard dummy text ever since the 1500s, when an unknown printer took a galley of type and scrambled it to make a type specimen book. It has survived not only five centuries, but also the leap into electronic typesetting, remaining essentially unchanged. It was popularised in the 1960s with the release of Letraset sheets containing Lorem Ipsum passages, and more recently with desktop publishing software like Aldus PageMaker including versions of Lorem Ipsum.
+Lorem Ipsum is simply dummy text of the printing and typesetting industry.
+Lorem Ipsum has been the industry's standard dummy text ever since the 1500s,
+when an unknown printer took a galley of type and scrambled it to make a type
+specimen book. It has survived not only five centuries, but also the leap into
+electronic typesetting, remaining essentially unchanged. It was popularised in
+the 1960s with the release of Letraset sheets containing Lorem Ipsum passages,
+and more recently with desktop publishing software like Aldus PageMaker
+including versions of Lorem Ipsum.
 
 
 
@@ -27,10 +34,9 @@ Lorem Ipsum is simply dummy text of the printing and typesetting industry. Lorem
 
 
 ```python
-import sys
+"""
 
-def greeting():
-    print("Hello world!")
+_text_suffix = """
 ```
 
 ## Sub header too
@@ -41,81 +47,146 @@ The end.
 
 
 class MarkdownStream:
-    live = None
-    when = 0
-    min_delay = 0.050
-    live_window = 6
+    """Streaming markdown renderer that progressively displays content with a live updating window.
+
+    Uses rich.console and rich.live to render markdown content with smooth scrolling
+    and partial updates. Maintains a sliding window of visible content while streaming
+    in new markdown text.
+    """
+
+    live = None  # Rich Live display instance
+    when = 0  # Timestamp of last update
+    min_delay = 1.0 / 20  # Minimum time between updates (20fps)
+    live_window = 6  # Number of lines to keep visible at bottom during streaming
 
     def __init__(self, mdargs=None):
-        self.printed = []
+        """Initialize the markdown stream.
+
+        Args:
+            mdargs (dict, optional): Additional arguments to pass to rich Markdown renderer
+        """
+        self.printed = []  # Stores lines that have already been printed
 
         if mdargs:
             self.mdargs = mdargs
         else:
             self.mdargs = dict()
 
+        # Initialize rich Live display with empty text
         self.live = Live(Text(""), refresh_per_second=1.0 / self.min_delay)
         self.live.start()
 
+    def _render_markdown_to_lines(self, text):
+        """Render markdown text to a list of lines.
+
+        Args:
+            text (str): Markdown text to render
+
+        Returns:
+            list: List of rendered lines with line endings preserved
+        """
+        # Render the markdown to a string buffer
+        string_io = io.StringIO()
+        console = Console(file=string_io, force_terminal=True)
+        markdown = Markdown(text, **self.mdargs)
+        console.print(markdown)
+        output = string_io.getvalue()
+
+        # Split rendered output into lines
+        return output.splitlines(keepends=True)
+
     def __del__(self):
+        """Destructor to ensure Live display is properly cleaned up."""
         if self.live:
             try:
                 self.live.stop()
             except Exception:
-                pass
+                pass  # Ignore any errors during cleanup
 
     def update(self, text, final=False):
+        """Update the displayed markdown content.
+
+        Args:
+            text (str): The markdown text received so far
+            final (bool): If True, this is the final update and we should clean up
+
+        Splits the output into "stable" older lines and the "last few" lines
+        which aren't considered stable. They may shift around as new chunks
+        are appended to the markdown text.
+
+        The stable lines emit to the console above the Live window.
+        The unstable lines emit into the Live window so they can be repainted.
+
+        Markdown going to the console works better in terminal scrollback buffers.
+        The live window doesn't play nice with terminal scrollback.
+        """
         now = time.time()
+        # Throttle updates to maintain smooth rendering
         if not final and now - self.when < self.min_delay:
             return
         self.when = now
 
-        string_io = io.StringIO()
-        console = Console(file=string_io, force_terminal=True)
+        # Measure render time and adjust min_delay to maintain smooth rendering
+        start = time.time()
+        lines = self._render_markdown_to_lines(text)
+        render_time = time.time() - start
 
-        markdown = Markdown(text, **self.mdargs)
+        # Set min_delay to render time plus a small buffer
+        self.min_delay = min(max(render_time * 10, 1.0 / 20), 2)
 
-        console.print(markdown)
-        output = string_io.getvalue()
-
-        lines = output.splitlines(keepends=True)
         num_lines = len(lines)
 
+        # How many lines have "left" the live window and are now considered stable?
+        # Or if final, consider all lines to be stable.
         if not final:
             num_lines -= self.live_window
 
+        # If we have stable content to display...
         if final or num_lines > 0:
+            # How many stable lines do we need to newly show above the live window?
             num_printed = len(self.printed)
-
             show = num_lines - num_printed
 
+            # Skip if no new lines to show above live window
             if show <= 0:
                 return
 
+            # Get the new lines and display them
             show = lines[num_printed:num_lines]
             show = "".join(show)
             show = Text.from_ansi(show)
-            self.live.console.print(show)
+            self.live.console.print(show)  # to the console above the live area
 
+            # Update our record of printed lines
             self.printed = lines[:num_lines]
 
+        # Handle final update cleanup
         if final:
             self.live.update(Text(""))
             self.live.stop()
             self.live = None
-        else:
-            rest = lines[num_lines:]
-            rest = "".join(rest)
-            # rest = '...\n' + rest
-            rest = Text.from_ansi(rest)
-            self.live.update(rest)
+            return
+
+        # Update the live window with remaining lines
+        rest = lines[num_lines:]
+        rest = "".join(rest)
+        rest = Text.from_ansi(rest)
+        self.live.update(rest)
+
+    def find_minimal_suffix(self, text, match_lines=50):
+        """
+        Splits text into chunks on blank lines "\n\n".
+        """
 
 
 if __name__ == "__main__":
-    _text = 5 * _text
+    with open("aider/io.py", "r") as f:
+        code = f.read()
+    _text = _text_prefix + code + _text_suffix
+    _text = _text * 10
 
     pm = MarkdownStream()
-    for i in range(6, len(_text)):
+    for i in range(6, len(_text), 5):
         pm.update(_text[:i])
         time.sleep(0.01)
 

aider/queries/tree-sitter-kotlin-tags.scm

@@ -0,0 +1,27 @@
+; Definitions
+
+(class_declaration
+  (type_identifier) @name.definition.class) @definition.class
+
+(function_declaration
+  (simple_identifier) @name.definition.function) @definition.function
+
+(object_declaration
+  (type_identifier) @name.definition.object) @definition.object
+
+; References
+
+(call_expression
+  [
+    (simple_identifier) @name.reference.call
+    (navigation_expression
+      (navigation_suffix
+        (simple_identifier) @name.reference.call))
+  ]) @reference.call
+
+(delegation_specifier
+  [
+    (user_type) @name.reference.type
+    (constructor_invocation
+      (user_type) @name.reference.type)
+  ]) @reference.type

aider/repo.py

@@ -2,23 +2,35 @@ import os
 import time
 from pathlib import Path, PurePosixPath
 
-import git
+try:
+    import git
+
+    ANY_GIT_ERROR = [
+        git.exc.ODBError,
+        git.exc.GitError,
+        git.exc.InvalidGitRepositoryError,
+    ]
+except ImportError:
+    git = None
+    ANY_GIT_ERROR = []
+
 import pathspec
 
 from aider import prompts, utils
-from aider.sendchat import simple_send_with_retries
 
 from .dump import dump  # noqa: F401
 
-ANY_GIT_ERROR = (
-    git.exc.ODBError,
-    git.exc.GitError,
+ANY_GIT_ERROR += [
     OSError,
     IndexError,
     BufferError,
     TypeError,
     ValueError,
-)
+    AttributeError,
+    AssertionError,
+    TimeoutError,
+]
+ANY_GIT_ERROR = tuple(ANY_GIT_ERROR)
 
 
 class GitRepo:
@@ -141,7 +153,7 @@ class GitRepo:
             os.environ["GIT_COMMITTER_NAME"] = committer_name
 
         if aider_edits and self.attribute_author:
-            original_auther_name_env = os.environ.get("GIT_AUTHOR_NAME")
+            original_author_name_env = os.environ.get("GIT_AUTHOR_NAME")
             os.environ["GIT_AUTHOR_NAME"] = committer_name
 
         try:
@@ -161,8 +173,8 @@ class GitRepo:
                     del os.environ["GIT_COMMITTER_NAME"]
 
             if aider_edits and self.attribute_author:
-                if original_auther_name_env is not None:
-                    os.environ["GIT_AUTHOR_NAME"] = original_auther_name_env
+                if original_author_name_env is not None:
+                    os.environ["GIT_AUTHOR_NAME"] = original_author_name_env
                 else:
                     del os.environ["GIT_AUTHOR_NAME"]
 
@@ -192,7 +204,7 @@ class GitRepo:
             max_tokens = model.info.get("max_input_tokens") or 0
             if max_tokens and num_tokens > max_tokens:
                 continue
-            commit_message = simple_send_with_retries(model, messages)
+            commit_message = model.simple_send_with_retries(messages)
             if commit_message:
                 break
 
@@ -276,9 +288,17 @@ class GitRepo:
                 files = self.tree_files[commit]
             else:
                 try:
-                    for blob in commit.tree.traverse():
-                        if blob.type == "blob":  # blob is a file
-                            files.add(blob.path)
+                    iterator = commit.tree.traverse()
+                    while True:
+                        try:
+                            blob = next(iterator)
+                            if blob.type == "blob":  # blob is a file
+                                files.add(blob.path)
+                        except IndexError:
+                            self.io.tool_warning(f"GitRepo: read error skipping {blob.path}")
+                            continue
+                        except StopIteration:
+                            break
                 except ANY_GIT_ERROR as err:
                     self.git_repo_error = err
                     self.io.tool_error(f"Unable to list files in git repo: {err}")
@@ -350,8 +370,8 @@ class GitRepo:
 
     def ignored_file_raw(self, fname):
         if self.subtree_only:
-            fname_path = Path(self.normalize_path(fname))
             try:
+                fname_path = Path(self.normalize_path(fname))
                 cwd_path = Path.cwd().resolve().relative_to(Path(self.root).resolve())
             except ValueError:
                 # Issue #1524

aider/repomap.py

@@ -605,7 +605,7 @@ class RepoMap:
 
         self.tree_cache = dict()
 
-        middle = min(max_map_tokens // 25, num_tags)
+        middle = min(int(max_map_tokens // 25), num_tags)
         while lower_bound <= upper_bound:
             # dump(lower_bound, middle, upper_bound)
 
@@ -628,7 +628,7 @@ class RepoMap:
             else:
                 upper_bound = middle - 1
 
-            middle = (lower_bound + upper_bound) // 2
+            middle = int((lower_bound + upper_bound) // 2)
 
         spin.end()
         return best_tree

aider/resources/model-metadata.json

@@ -0,0 +1,118 @@
+{
+    "deepseek-reasoner": {
+        "max_tokens": 8192,
+        "max_input_tokens": 64000,
+        "max_output_tokens": 8192,
+        "input_cost_per_token": 0.00000055,
+        "input_cost_per_token_cache_hit": 0.00000014,
+        "cache_read_input_token_cost": 0.00000014,
+        "cache_creation_input_token_cost": 0.0,
+        "output_cost_per_token": 0.00000219,
+        "litellm_provider": "deepseek",
+        "mode": "chat",
+        //"supports_function_calling": true, 
+        "supports_assistant_prefill": true,
+        //"supports_tool_choice": true,
+        "supports_prompt_caching": true
+    },
+    "openrouter/deepseek/deepseek-r1": {
+        "max_tokens": 8192,
+        "max_input_tokens": 64000,
+        "max_output_tokens": 8192,
+        "input_cost_per_token": 0.00000055,
+        "input_cost_per_token_cache_hit": 0.00000014,
+        "cache_read_input_token_cost": 0.00000014,
+        "cache_creation_input_token_cost": 0.0,
+        "output_cost_per_token": 0.00000219,
+        "litellm_provider": "openrouter",
+        "mode": "chat",
+        //"supports_function_calling": true, 
+        "supports_assistant_prefill": true,
+        //"supports_tool_choice": true,
+        "supports_prompt_caching": true
+    },
+    "openrouter/deepseek/deepseek-r1:free": {
+        "max_tokens": 8192,
+        "max_input_tokens": 64000,
+        "max_output_tokens": 8192,
+        "input_cost_per_token": 0.0,
+        "input_cost_per_token_cache_hit": 0.0,
+        "cache_read_input_token_cost": 0.00,
+        "cache_creation_input_token_cost": 0.0,
+        "output_cost_per_token": 0.0,
+        "litellm_provider": "openrouter",
+        "mode": "chat",
+        //"supports_function_calling": true, 
+        "supports_assistant_prefill": true,
+        //"supports_tool_choice": true,
+        "supports_prompt_caching": true
+    },
+    "fireworks_ai/accounts/fireworks/models/deepseek-r1": {
+        "max_tokens": 160000,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 20480,
+        "litellm_provider": "fireworks_ai",
+        "input_cost_per_token": 0.000008,
+        "output_cost_per_token": 0.000008,
+        "mode": "chat",
+    },
+    "fireworks_ai/accounts/fireworks/models/deepseek-v3": {
+        "max_tokens": 128000,
+        "max_input_tokens": 100000,
+        "max_output_tokens": 8192,
+        "litellm_provider": "fireworks_ai",
+        "input_cost_per_token": 0.0000009,
+        "output_cost_per_token": 0.0000009,
+        "mode": "chat",
+    },
+    "o3-mini": {
+        "max_tokens": 100000,
+        "max_input_tokens": 200000,
+        "max_output_tokens": 100000,
+        "input_cost_per_token": 0.0000011,
+        "output_cost_per_token": 0.0000044,
+        "cache_read_input_token_cost": 0.00000055,
+        "litellm_provider": "openai",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_vision": true,
+        "supports_prompt_caching": true,
+        "supports_system_messages": true,
+        "supports_response_schema": true
+    },
+    "openrouter/openai/o3-mini": {
+        "max_tokens": 100000,
+        "max_input_tokens": 200000,
+        "max_output_tokens": 100000,
+        "input_cost_per_token": 0.0000011,
+        "output_cost_per_token": 0.0000044,
+        "cache_read_input_token_cost": 0.00000055,
+        "litellm_provider": "openrouter",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_vision": true,
+        "supports_prompt_caching": true,
+        "supports_system_messages": true,
+        "supports_response_schema": true
+    },
+    "openrouter/openai/gpt-4o-mini": {
+        "max_tokens": 16384,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 16384,
+        "input_cost_per_token": 0.00000015,
+        "output_cost_per_token": 0.00000060,
+        "input_cost_per_token_batches": 0.000000075,
+        "output_cost_per_token_batches": 0.00000030,
+        "cache_read_input_token_cost": 0.000000075,
+        "litellm_provider": "openrouter",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_response_schema": true,
+        "supports_vision": true,
+        "supports_prompt_caching": true,
+        "supports_system_messages": true
+    },
+}

aider/resources/model-settings.yml

@@ -0,0 +1,662 @@
+- name: gpt-3.5-turbo
+  weak_model_name: gpt-4o-mini
+  reminder: sys
+
+- name: gpt-3.5-turbo-0125
+  weak_model_name: gpt-4o-mini
+  reminder: sys
+
+- name: gpt-3.5-turbo-1106
+  weak_model_name: gpt-4o-mini
+  reminder: sys
+
+- name: gpt-3.5-turbo-0613
+  weak_model_name: gpt-4o-mini
+  reminder: sys
+
+- name: gpt-3.5-turbo-16k-0613
+  weak_model_name: gpt-4o-mini
+  reminder: sys
+
+- name: gpt-4-turbo-2024-04-09
+  edit_format: udiff
+  weak_model_name: gpt-4o-mini
+  use_repo_map: true
+  lazy: true
+  reminder: sys
+
+- name: gpt-4-turbo
+  edit_format: udiff
+  weak_model_name: gpt-4o-mini
+  use_repo_map: true
+  lazy: true
+  reminder: sys
+
+- name: openai/gpt-4o
+  edit_format: diff
+  weak_model_name: gpt-4o-mini
+  use_repo_map: true
+  lazy: true
+  reminder: sys
+  examples_as_sys_msg: true
+  editor_edit_format: editor-diff
+
+- name: openai/gpt-4o-2024-08-06
+  edit_format: diff
+  weak_model_name: gpt-4o-mini
+  use_repo_map: true
+  lazy: true
+  reminder: sys
+  examples_as_sys_msg: true
+
+- name: gpt-4o-2024-08-06
+  edit_format: diff
+  weak_model_name: gpt-4o-mini
+  use_repo_map: true
+  lazy: true
+  reminder: sys
+  examples_as_sys_msg: true
+
+- name: gpt-4o-2024-11-20
+  edit_format: diff
+  weak_model_name: gpt-4o-mini
+  use_repo_map: true
+  lazy: true
+  reminder: sys
+  examples_as_sys_msg: true
+
+- name: openai/gpt-4o-2024-11-20
+  edit_format: diff
+  weak_model_name: gpt-4o-mini
+  use_repo_map: true
+  lazy: true
+  reminder: sys
+  examples_as_sys_msg: true
+
+- name: gpt-4o
+  edit_format: diff
+  weak_model_name: gpt-4o-mini
+  use_repo_map: true
+  lazy: true
+  reminder: sys
+  examples_as_sys_msg: true
+  editor_edit_format: editor-diff
+
+- name: gpt-4o-mini
+  weak_model_name: gpt-4o-mini
+  lazy: true
+  reminder: sys
+
+- name: openai/gpt-4o-mini
+  weak_model_name: openai/gpt-4o-mini
+  lazy: true
+  reminder: sys
+
+- name: gpt-4-0125-preview
+  edit_format: udiff
+  weak_model_name: gpt-4o-mini
+  use_repo_map: true
+  lazy: true
+  reminder: sys
+  examples_as_sys_msg: true
+
+- name: gpt-4-1106-preview
+  edit_format: udiff
+  weak_model_name: gpt-4o-mini
+  use_repo_map: true
+  lazy: true
+  reminder: sys
+
+- name: gpt-4-vision-preview
+  edit_format: diff
+  weak_model_name: gpt-4o-mini
+  use_repo_map: true
+  reminder: sys
+
+- name: gpt-4-0314
+  edit_format: diff
+  weak_model_name: gpt-4o-mini
+  use_repo_map: true
+  reminder: sys
+  examples_as_sys_msg: true
+
+- name: gpt-4-0613
+  edit_format: diff
+  weak_model_name: gpt-4o-mini
+  use_repo_map: true
+  reminder: sys
+
+- name: gpt-4-32k-0613
+  edit_format: diff
+  weak_model_name: gpt-4o-mini
+  use_repo_map: true
+  reminder: sys
+
+- name: claude-3-opus-20240229
+  edit_format: diff
+  weak_model_name: claude-3-5-haiku-20241022
+  use_repo_map: true
+
+- name: openrouter/anthropic/claude-3-opus
+  edit_format: diff
+  weak_model_name: openrouter/anthropic/claude-3-5-haiku
+  use_repo_map: true
+
+- name: claude-3-sonnet-20240229
+  weak_model_name: claude-3-5-haiku-20241022
+
+- name: claude-3-5-sonnet-20240620
+  edit_format: diff
+  weak_model_name: claude-3-5-haiku-20241022
+  use_repo_map: true
+  examples_as_sys_msg: true
+  extra_params:
+    extra_headers:
+      anthropic-beta: prompt-caching-2024-07-31,pdfs-2024-09-25
+    max_tokens: 8192
+  cache_control: true
+  editor_model_name: claude-3-5-sonnet-20240620
+  editor_edit_format: editor-diff
+
+- name: anthropic/claude-3-5-sonnet-20240620
+  edit_format: diff
+  weak_model_name: anthropic/claude-3-5-haiku-20241022
+  use_repo_map: true
+  examples_as_sys_msg: true
+  extra_params:
+    extra_headers:
+      anthropic-beta: prompt-caching-2024-07-31,pdfs-2024-09-25
+    max_tokens: 8192
+  cache_control: true
+  editor_model_name: anthropic/claude-3-5-sonnet-20240620
+  editor_edit_format: editor-diff
+
+- name: anthropic/claude-3-5-sonnet-20241022
+  edit_format: diff
+  weak_model_name: anthropic/claude-3-5-haiku-20241022
+  use_repo_map: true
+  examples_as_sys_msg: true
+  extra_params:
+    extra_headers:
+      anthropic-beta: prompt-caching-2024-07-31,pdfs-2024-09-25
+    max_tokens: 8192
+  cache_control: true
+  editor_model_name: anthropic/claude-3-5-sonnet-20241022
+  editor_edit_format: editor-diff
+
+- name: bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0
+  edit_format: diff
+  weak_model_name: bedrock/anthropic.claude-3-5-haiku-20241022-v1:0
+  use_repo_map: true
+  examples_as_sys_msg: true
+  extra_params:
+    extra_headers:
+      anthropic-beta: prompt-caching-2024-07-31,pdfs-2024-09-25
+    max_tokens: 8192
+  cache_control: true
+  editor_model_name: bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0
+  editor_edit_format: editor-diff
+
+- name: anthropic/claude-3-5-sonnet-latest
+  edit_format: diff
+  weak_model_name: anthropic/claude-3-5-haiku-20241022
+  use_repo_map: true
+  examples_as_sys_msg: true
+  extra_params:
+    extra_headers:
+      anthropic-beta: prompt-caching-2024-07-31,pdfs-2024-09-25
+    max_tokens: 8192
+  cache_control: true
+  editor_model_name: anthropic/claude-3-5-sonnet-20241022
+  editor_edit_format: editor-diff
+
+- name: claude-3-5-sonnet-20241022
+  edit_format: diff
+  weak_model_name: claude-3-5-haiku-20241022
+  use_repo_map: true
+  examples_as_sys_msg: true
+  extra_params:
+    extra_headers:
+      anthropic-beta: prompt-caching-2024-07-31,pdfs-2024-09-25
+    max_tokens: 8192
+  cache_control: true
+  editor_model_name: claude-3-5-sonnet-20241022
+  editor_edit_format: editor-diff
+
+- name: anthropic/claude-3-haiku-20240307
+  weak_model_name: anthropic/claude-3-haiku-20240307
+  examples_as_sys_msg: true
+  extra_params:
+    extra_headers:
+      anthropic-beta: prompt-caching-2024-07-31,pdfs-2024-09-25
+  cache_control: true
+
+- name: anthropic/claude-3-5-haiku-20241022
+  edit_format: diff
+  weak_model_name: anthropic/claude-3-5-haiku-20241022
+  use_repo_map: true
+  extra_params:
+    extra_headers:
+      anthropic-beta: prompt-caching-2024-07-31,pdfs-2024-09-25
+  cache_control: true
+
+- name: bedrock/anthropic.claude-3-5-haiku-20241022-v1:0
+  edit_format: diff
+  weak_model_name: bedrock/anthropic.claude-3-5-haiku-20241022-v1:0
+  use_repo_map: true
+  extra_params:
+    extra_headers:
+      anthropic-beta: prompt-caching-2024-07-31,pdfs-2024-09-25
+  cache_control: true
+
+- name: claude-3-5-haiku-20241022
+  edit_format: diff
+  weak_model_name: claude-3-5-haiku-20241022
+  use_repo_map: true
+  examples_as_sys_msg: true
+  extra_params:
+    extra_headers:
+      anthropic-beta: prompt-caching-2024-07-31,pdfs-2024-09-25
+  cache_control: true
+
+- name: vertex_ai/claude-3-5-haiku@20241022
+  edit_format: diff
+  weak_model_name: vertex_ai/claude-3-5-haiku@20241022
+  use_repo_map: true
+  extra_params:
+    max_tokens: 4096
+
+- name: claude-3-haiku-20240307
+  weak_model_name: claude-3-haiku-20240307
+  examples_as_sys_msg: true
+  extra_params:
+    extra_headers:
+      anthropic-beta: prompt-caching-2024-07-31,pdfs-2024-09-25
+  cache_control: true
+
+- name: openrouter/anthropic/claude-3.5-sonnet
+  edit_format: diff
+  weak_model_name: openrouter/anthropic/claude-3-5-haiku
+  use_repo_map: true
+  examples_as_sys_msg: true
+  extra_params:
+    max_tokens: 8192
+  cache_control: true
+  editor_model_name: openrouter/anthropic/claude-3.5-sonnet
+  editor_edit_format: editor-diff
+
+- name: openrouter/anthropic/claude-3.5-sonnet:beta
+  edit_format: diff
+  weak_model_name: openrouter/anthropic/claude-3-5-haiku:beta
+  use_repo_map: true
+  examples_as_sys_msg: true
+  extra_params:
+    max_tokens: 8192
+  cache_control: true
+  editor_model_name: openrouter/anthropic/claude-3.5-sonnet:beta
+  editor_edit_format: editor-diff
+
+- name: vertex_ai/claude-3-5-sonnet@20240620
+  edit_format: diff
+  weak_model_name: vertex_ai/claude-3-5-haiku@20241022
+  use_repo_map: true
+  examples_as_sys_msg: true
+  extra_params:
+    max_tokens: 8192
+  editor_model_name: vertex_ai/claude-3-5-sonnet@20240620
+  editor_edit_format: editor-diff
+
+- name: vertex_ai/claude-3-5-sonnet-v2@20241022
+  edit_format: diff
+  weak_model_name: vertex_ai/claude-3-5-haiku@20241022
+  use_repo_map: true
+  examples_as_sys_msg: true
+  extra_params:
+    max_tokens: 8192
+  editor_model_name: vertex_ai/claude-3-5-sonnet-v2@20241022
+  editor_edit_format: editor-diff
+
+- name: vertex_ai/claude-3-opus@20240229
+  edit_format: diff
+  weak_model_name: vertex_ai/claude-3-5-haiku@20241022
+  use_repo_map: true
+
+- name: vertex_ai/claude-3-sonnet@20240229
+  weak_model_name: vertex_ai/claude-3-5-haiku@20241022
+
+- name: command-r-plus
+  weak_model_name: command-r-plus
+  use_repo_map: true
+
+- name: command-r-08-2024
+  weak_model_name: command-r-08-2024
+  use_repo_map: true
+
+- name: command-r-plus-08-2024
+  weak_model_name: command-r-plus-08-2024
+  use_repo_map: true
+
+- name: groq/llama3-70b-8192
+  edit_format: diff
+  weak_model_name: groq/llama3-8b-8192
+  examples_as_sys_msg: true
+
+- name: openrouter/meta-llama/llama-3-70b-instruct
+  edit_format: diff
+  weak_model_name: openrouter/meta-llama/llama-3-70b-instruct
+  examples_as_sys_msg: true
+
+- name: gemini/gemini-1.5-pro-002
+  edit_format: diff
+  use_repo_map: true
+
+- name: gemini/gemini-1.5-flash-002
+
+- name: gemini/gemini-1.5-pro
+  edit_format: diff-fenced
+  use_repo_map: true
+
+- name: gemini/gemini-1.5-pro-latest
+  edit_format: diff-fenced
+  use_repo_map: true
+
+- name: gemini/gemini-1.5-pro-exp-0827
+  edit_format: diff-fenced
+  use_repo_map: true
+
+- name: gemini/gemini-exp-1206
+  edit_format: diff
+  use_repo_map: true
+
+- name: gemini/gemini-exp-1114
+  edit_format: diff
+  use_repo_map: true
+
+- name: gemini/gemini-exp-1121
+  edit_format: diff
+  use_repo_map: true
+
+- name: vertex_ai/gemini-pro-experimental
+  edit_format: diff-fenced
+  use_repo_map: true
+
+- name: gemini/gemini-1.5-flash-exp-0827
+
+- name: gemini/gemini-2.0-flash-exp
+  edit_format: diff
+  use_repo_map: true
+
+- name: gemini/gemini-2.0-flash
+  edit_format: diff
+  use_repo_map: true
+
+- name: openrouter/deepseek/deepseek-r1
+  edit_format: diff
+  weak_model_name: openrouter/deepseek/deepseek-chat
+  use_repo_map: true
+  examples_as_sys_msg: true
+  extra_params:
+    max_tokens: 8192
+  caches_by_default: true
+  use_temperature: false
+  editor_model_name: openrouter/deepseek/deepseek-chat
+  editor_edit_format: editor-diff
+
+- name: openrouter/deepseek/deepseek-r1:free
+  edit_format: diff
+  weak_model_name: openrouter/deepseek/deepseek-r1:free
+  use_repo_map: true
+  examples_as_sys_msg: true
+  extra_params:
+    max_tokens: 8192
+  caches_by_default: true
+  use_temperature: false
+  editor_model_name: openrouter/deepseek/deepseek-r1:free
+  editor_edit_format: editor-diff
+
+- name: deepseek/deepseek-reasoner
+  edit_format: diff
+  weak_model_name: deepseek/deepseek-chat
+  use_repo_map: true
+  examples_as_sys_msg: true
+  extra_params:
+    max_tokens: 8192
+  caches_by_default: true
+  use_temperature: false
+  editor_model_name: deepseek/deepseek-chat
+  editor_edit_format: editor-diff
+
+- name: deepseek/deepseek-chat
+  edit_format: diff
+  use_repo_map: true
+  reminder: sys
+  examples_as_sys_msg: true
+  extra_params:
+    max_tokens: 8192
+  caches_by_default: true
+
+- name: deepseek/deepseek-coder
+  edit_format: diff
+  use_repo_map: true
+  reminder: sys
+  examples_as_sys_msg: true
+  extra_params:
+    max_tokens: 8192
+  caches_by_default: true
+
+- name: deepseek-chat
+  edit_format: diff
+  use_repo_map: true
+  reminder: sys
+  examples_as_sys_msg: true
+  extra_params:
+    max_tokens: 8192
+
+- name: deepseek-coder
+  edit_format: diff
+  use_repo_map: true
+  reminder: sys
+  examples_as_sys_msg: true
+  extra_params:
+    max_tokens: 8192
+  caches_by_default: true
+
+- name: openrouter/deepseek/deepseek-coder
+  edit_format: diff
+  use_repo_map: true
+  reminder: sys
+  examples_as_sys_msg: true
+
+- name: openrouter/deepseek/deepseek-chat
+  edit_format: diff
+  use_repo_map: true
+  reminder: sys
+  examples_as_sys_msg: true
+
+- name: openrouter/openai/gpt-4o
+  edit_format: diff
+  weak_model_name: openrouter/openai/gpt-4o-mini
+  use_repo_map: true
+  lazy: true
+  reminder: sys
+  examples_as_sys_msg: true
+  editor_edit_format: editor-diff
+
+- name: openai/o1-mini
+  weak_model_name: openai/gpt-4o-mini
+  use_repo_map: true
+  use_system_prompt: false
+  use_temperature: false
+  editor_model_name: openai/gpt-4o
+  editor_edit_format: editor-diff
+
+- name: azure/o1-mini
+  weak_model_name: azure/gpt-4o-mini
+  use_repo_map: true
+  use_system_prompt: false
+  use_temperature: false
+  editor_model_name: azure/gpt-4o
+  editor_edit_format: editor-diff
+
+- name: o1-mini
+  weak_model_name: gpt-4o-mini
+  use_repo_map: true
+  use_system_prompt: false
+  use_temperature: false
+  editor_model_name: gpt-4o
+  editor_edit_format: editor-diff
+
+- name: openai/o1-preview
+  edit_format: diff
+  weak_model_name: openai/gpt-4o-mini
+  use_repo_map: true
+  use_system_prompt: false
+  use_temperature: false
+  editor_model_name: openai/gpt-4o
+  editor_edit_format: editor-diff
+
+- name: azure/o1-preview
+  edit_format: diff
+  weak_model_name: azure/gpt-4o-mini
+  use_repo_map: true
+  use_system_prompt: false
+  use_temperature: false
+  editor_model_name: azure/gpt-4o
+  editor_edit_format: editor-diff
+
+- name: azure/o1
+  edit_format: diff
+  weak_model_name: azure/gpt-4o-mini
+  use_repo_map: true
+  use_temperature: false
+  streaming: false
+  editor_model_name: azure/gpt-4o
+  editor_edit_format: editor-diff
+
+- name: o1-preview
+  edit_format: architect
+  weak_model_name: gpt-4o-mini
+  use_repo_map: true
+  use_system_prompt: false
+  use_temperature: false
+  editor_model_name: gpt-4o
+  editor_edit_format: editor-diff
+
+- name: openrouter/openai/o1-mini
+  weak_model_name: openrouter/openai/gpt-4o-mini
+  use_repo_map: true
+  use_system_prompt: false
+  use_temperature: false
+  streaming: false
+  editor_model_name: openrouter/openai/gpt-4o
+  editor_edit_format: editor-diff
+
+- name: openrouter/openai/o1-preview
+  edit_format: diff
+  weak_model_name: openrouter/openai/gpt-4o-mini
+  use_repo_map: true
+  use_system_prompt: false
+  use_temperature: false
+  streaming: false
+  editor_model_name: openrouter/openai/gpt-4o
+  editor_edit_format: editor-diff
+
+- name: openrouter/openai/o1
+  edit_format: diff
+  weak_model_name: openrouter/openai/gpt-4o-mini
+  use_repo_map: true
+  use_temperature: false
+  streaming: false
+  editor_model_name: openrouter/openai/gpt-4o
+  editor_edit_format: editor-diff
+
+- name: openai/o1
+  edit_format: diff
+  weak_model_name: openai/gpt-4o-mini
+  use_repo_map: true
+  use_temperature: false
+  streaming: false
+  editor_model_name: openai/gpt-4o
+  editor_edit_format: editor-diff
+
+- name: o1
+  edit_format: diff
+  weak_model_name: gpt-4o-mini
+  use_repo_map: true
+  use_temperature: false
+  streaming: false
+  editor_model_name: gpt-4o
+  editor_edit_format: editor-diff
+
+- name: openrouter/qwen/qwen-2.5-coder-32b-instruct
+  edit_format: diff
+  weak_model_name: openrouter/qwen/qwen-2.5-coder-32b-instruct
+  use_repo_map: true
+  editor_model_name: openrouter/qwen/qwen-2.5-coder-32b-instruct
+  editor_edit_format: editor-diff
+
+- name: openrouter/deepseek/deepseek-r1-distill-llama-70b
+  edit_format: diff
+  weak_model_name: openrouter/deepseek/deepseek-chat
+  use_repo_map: true
+  examples_as_sys_msg: true
+  extra_params:
+    max_tokens: 8192
+  caches_by_default: true
+  use_temperature: false
+  editor_model_name: openrouter/deepseek/deepseek-chat
+  editor_edit_format: editor-diff
+  
+- name: fireworks_ai/accounts/fireworks/models/deepseek-r1
+  edit_format: diff
+  weak_model_name: fireworks_ai/accounts/fireworks/models/deepseek-v3
+  use_repo_map: true
+  use_temperature: false
+  streaming: true
+  editor_model_name: fireworks_ai/accounts/fireworks/models/deepseek-v3
+  editor_edit_format: editor-diff
+  remove_reasoning: think
+  extra_params:
+      max_tokens: 160000
+
+- name: fireworks_ai/accounts/fireworks/models/deepseek-v3
+  edit_format: diff
+  use_repo_map: true
+  reminder: sys
+  examples_as_sys_msg: true
+  extra_params:
+      max_tokens: 128000
+
+- name: openai/o3-mini
+  edit_format: diff
+  weak_model_name: gpt-4o-mini
+  use_repo_map: true
+  use_temperature: false
+  editor_model_name: gpt-4o
+  editor_edit_format: editor-diff
+  
+- name: o3-mini
+  edit_format: diff
+  weak_model_name: gpt-4o-mini
+  use_repo_map: true
+  use_temperature: false
+  editor_model_name: gpt-4o
+  editor_edit_format: editor-diff
+
+- name: openrouter/openai/o3-mini
+  edit_format: diff
+  weak_model_name: openrouter/openai/gpt-4o-mini
+  use_repo_map: true
+  use_temperature: false
+  editor_model_name: openrouter/openai/gpt-4o
+  editor_edit_format: editor-diff
+
+- name: azure/o3-mini
+  edit_format: diff
+  weak_model_name: azure/gpt-4o-mini
+  use_repo_map: true
+  use_temperature: false
+  editor_model_name: azure/gpt-4o
+  editor_edit_format: editor-diff
+
+  

aider/run_cmd.py

@@ -39,7 +39,7 @@ def get_windows_parent_process_name():
         return None
 
 
-def run_cmd_subprocess(command, verbose=False, cwd=None):
+def run_cmd_subprocess(command, verbose=False, cwd=None, encoding=sys.stdout.encoding):
     if verbose:
         print("Using run_cmd_subprocess:", command)
 
@@ -65,7 +65,7 @@ def run_cmd_subprocess(command, verbose=False, cwd=None):
             stderr=subprocess.STDOUT,
             text=True,
             shell=True,
-            encoding=sys.stdout.encoding,
+            encoding=encoding,
             errors="replace",
             bufsize=0,  # Set bufsize to 0 for unbuffered output
             universal_newlines=True,

aider/sendchat.py

@@ -1,98 +1,61 @@
-import hashlib
-import json
-import time
-
 from aider.dump import dump  # noqa: F401
-from aider.exceptions import LiteLLMExceptions
-from aider.llm import litellm
-
-# from diskcache import Cache
+from aider.utils import format_messages
 
 
-CACHE_PATH = "~/.aider.send.cache.v1"
-CACHE = None
-# CACHE = Cache(CACHE_PATH)
+def sanity_check_messages(messages):
+    """Check if messages alternate between user and assistant roles.
+    System messages can be interspersed anywhere.
+    Also verifies the last non-system message is from the user.
+    Returns True if valid, False otherwise."""
+    last_role = None
+    last_non_system_role = None
 
-RETRY_TIMEOUT = 60
-
-
-def send_completion(
-    model_name,
-    messages,
-    functions,
-    stream,
-    temperature=0,
-    extra_params=None,
-):
-    kwargs = dict(
-        model=model_name,
-        messages=messages,
-        stream=stream,
-    )
-    if temperature is not None:
-        kwargs["temperature"] = temperature
-
-    if functions is not None:
-        function = functions[0]
-        kwargs["tools"] = [dict(type="function", function=function)]
-        kwargs["tool_choice"] = {"type": "function", "function": {"name": function["name"]}}
-
-    if extra_params is not None:
-        kwargs.update(extra_params)
-
-    key = json.dumps(kwargs, sort_keys=True).encode()
-
-    # Generate SHA1 hash of kwargs and append it to chat_completion_call_hashes
-    hash_object = hashlib.sha1(key)
-
-    if not stream and CACHE is not None and key in CACHE:
-        return hash_object, CACHE[key]
-
-    res = litellm.completion(**kwargs)
-
-    if not stream and CACHE is not None:
-        CACHE[key] = res
-
-    return hash_object, res
-
-
-def simple_send_with_retries(model, messages):
-    litellm_ex = LiteLLMExceptions()
-
-    retry_delay = 0.125
-    while True:
-        try:
-            kwargs = {
-                "model_name": model.name,
-                "messages": messages,
-                "functions": None,
-                "stream": False,
-                "temperature": None if not model.use_temperature else 0,
-                "extra_params": model.extra_params,
-            }
-
-            _hash, response = send_completion(**kwargs)
-            if not response or not hasattr(response, "choices") or not response.choices:
-                return None
-            return response.choices[0].message.content
-        except litellm_ex.exceptions_tuple() as err:
-            ex_info = litellm_ex.get_ex_info(err)
-
-            print(str(err))
-            if ex_info.description:
-                print(ex_info.description)
-
-            should_retry = ex_info.retry
-            if should_retry:
-                retry_delay *= 2
-                if retry_delay > RETRY_TIMEOUT:
-                    should_retry = False
-
-            if not should_retry:
-                return None
-
-            print(f"Retrying in {retry_delay:.1f} seconds...")
-            time.sleep(retry_delay)
+    for msg in messages:
+        role = msg.get("role")
+        if role == "system":
             continue
-        except AttributeError:
-            return None
+
+        if last_role and role == last_role:
+            turns = format_messages(messages)
+            raise ValueError("Messages don't properly alternate user/assistant:\n\n" + turns)
+
+        last_role = role
+        last_non_system_role = role
+
+    # Ensure last non-system message is from user
+    return last_non_system_role == "user"
+
+
+def ensure_alternating_roles(messages):
+    """Ensure messages alternate between 'assistant' and 'user' roles.
+
+    Inserts empty messages of the opposite role when consecutive messages
+    of the same role are found.
+
+    Args:
+        messages: List of message dictionaries with 'role' and 'content' keys.
+
+    Returns:
+        List of messages with alternating roles.
+    """
+    if not messages:
+        return messages
+
+    fixed_messages = []
+    prev_role = None
+
+    for msg in messages:
+        current_role = msg.get("role")  # Get 'role', None if missing
+
+        # If current role same as previous, insert empty message
+        # of the opposite role
+        if current_role == prev_role:
+            if current_role == "user":
+                fixed_messages.append({"role": "assistant", "content": ""})
+            else:
+                fixed_messages.append({"role": "user", "content": ""})
+
+        fixed_messages.append(msg)
+        prev_role = current_role
+
+    return fixed_messages

aider/urls.py

@@ -14,3 +14,4 @@ install_properly = "https://aider.chat/docs/troubleshooting/imports.html"
 analytics = "https://aider.chat/docs/more/analytics.html"
 release_notes = "https://aider.chat/HISTORY.html#release-notes"
 edit_formats = "https://aider.chat/docs/more/edit-formats.html"
+models_and_keys = "https://aider.chat/docs/troubleshooting/models-and-keys.html"

aider/utils.py

@@ -8,8 +8,6 @@ import tempfile
 import time
 from pathlib import Path
 
-import git
-
 from aider.dump import dump  # noqa: F401
 
 IMAGE_EXTENSIONS = {".png", ".jpg", ".jpeg", ".gif", ".bmp", ".tiff", ".webp", ".pdf"}
@@ -73,6 +71,8 @@ class GitTemporaryDirectory(ChdirTemporaryDirectory):
 
 
 def make_repo(path=None):
+    import git
+
     if not path:
         path = "."
     repo = git.Repo.init(path)
@@ -112,7 +112,7 @@ def format_messages(messages, title=None):
         output.append(f"{title.upper()} {'*' * 50}")
 
     for msg in messages:
-        output.append("")
+        output.append("-------")
         role = msg["role"].upper()
         content = msg.get("content")
         if isinstance(content, list):  # Handle list content (e.g., image messages)
@@ -300,12 +300,15 @@ class Spinner:
 
 
 def find_common_root(abs_fnames):
-    if len(abs_fnames) == 1:
-        return safe_abs_path(os.path.dirname(list(abs_fnames)[0]))
-    elif abs_fnames:
-        return safe_abs_path(os.path.commonpath(list(abs_fnames)))
-    else:
-        return safe_abs_path(os.getcwd())
+    try:
+        if len(abs_fnames) == 1:
+            return safe_abs_path(os.path.dirname(list(abs_fnames)[0]))
+        elif abs_fnames:
+            return safe_abs_path(os.path.commonpath(list(abs_fnames)))
+    except OSError:
+        pass
+
+    return safe_abs_path(os.getcwd())
 
 
 def format_tokens(count):

aider/voice.py

@@ -18,6 +18,7 @@ warnings.filterwarnings("ignore", category=SyntaxWarning)
 
 
 from pydub import AudioSegment  # noqa
+from pydub.exceptions import CouldntDecodeError, CouldntEncodeError  # noqa
 
 try:
     import soundfile as sf
@@ -140,13 +141,28 @@ class Voice:
             while not self.q.empty():
                 file.write(self.q.get())
 
-        if self.audio_format != "wav":
-            filename = tempfile.mktemp(suffix=f".{self.audio_format}")
-            audio = AudioSegment.from_wav(temp_wav)
-            audio.export(filename, format=self.audio_format)
-            os.remove(temp_wav)
-        else:
-            filename = temp_wav
+        use_audio_format = self.audio_format
+
+        # Check file size and offer to convert to mp3 if too large
+        file_size = os.path.getsize(temp_wav)
+        if file_size > 24.9 * 1024 * 1024 and self.audio_format == "wav":
+            print("\nWarning: {temp_wav} is too large, switching to mp3 format.")
+            use_audio_format = "mp3"
+
+        filename = temp_wav
+        if use_audio_format != "wav":
+            try:
+                new_filename = tempfile.mktemp(suffix=f".{use_audio_format}")
+                audio = AudioSegment.from_wav(temp_wav)
+                audio.export(new_filename, format=use_audio_format)
+                os.remove(temp_wav)
+                filename = new_filename
+            except (CouldntDecodeError, CouldntEncodeError) as e:
+                print(f"Error converting audio: {e}")
+            except (OSError, FileNotFoundError) as e:
+                print(f"File system error during conversion: {e}")
+            except Exception as e:
+                print(f"Unexpected error during audio conversion: {e}")
 
         with open(filename, "rb") as fh:
             try:
@@ -157,7 +173,7 @@ class Voice:
                 print(f"Unable to transcribe {filename}: {err}")
                 return
 
-        if self.audio_format != "wav":
+        if filename != temp_wav:
             os.remove(filename)
 
         text = transcript.text

aider/watch.py

@@ -9,59 +9,7 @@ from pathspec.patterns import GitWildMatchPattern
 from watchfiles import watch
 
 from aider.dump import dump  # noqa
-
-
-def is_source_file(path: Path) -> bool:
-    """
-    Check if a file is a source file that uses # or // style comments.
-    This includes Python, JavaScript, TypeScript, C, C++, etc.
-    """
-    COMMENT_STYLE_EXTENSIONS = {
-        # # style comments
-        ".py",
-        ".r",
-        ".rb",
-        ".pl",
-        ".pm",
-        ".sh",
-        ".bash",
-        ".zsh",
-        ".bashrc",
-        ".bash_profile",
-        ".bash_login",
-        ".bash_logout",
-        ".zshrc",
-        ".zprofile",
-        ".zlogin",
-        ".zlogout",
-        ".profile",
-        ".yaml",
-        ".yml",
-        # // style comments
-        ".js",
-        ".ts",
-        ".jsx",
-        ".tsx",
-        ".cpp",
-        ".c",
-        ".h",
-        ".hpp",
-        ".java",
-        ".swift",
-        ".kt",
-        ".cs",
-        ".go",
-        ".rs",
-        ".php",
-        # -- style comments
-        ".sql",
-        ".hs",  # Haskell
-        ".lua",
-        ".elm",
-        ".vhd",  # VHDL
-        ".vhdl",
-    }
-    return path.suffix.lower() in COMMENT_STYLE_EXTENSIONS
+from aider.watch_prompts import watch_ask_prompt, watch_code_prompt
 
 
 def load_gitignores(gitignore_paths: list[Path]) -> Optional[PathSpec]:
@@ -69,7 +17,41 @@ def load_gitignores(gitignore_paths: list[Path]) -> Optional[PathSpec]:
     if not gitignore_paths:
         return None
 
-    patterns = [".aider*", ".git"]  # Always ignore
+    patterns = [
+        ".aider*",
+        ".git",
+        # Common editor backup/temp files
+        "*~",  # Emacs/vim backup
+        "*.bak",  # Generic backup
+        "*.swp",  # Vim swap
+        "*.swo",  # Vim swap
+        "\\#*\\#",  # Emacs auto-save
+        ".#*",  # Emacs lock files
+        "*.tmp",  # Generic temp files
+        "*.temp",  # Generic temp files
+        "*.orig",  # Merge conflict originals
+        "*.pyc",  # Python bytecode
+        "__pycache__/",  # Python cache dir
+        ".DS_Store",  # macOS metadata
+        "Thumbs.db",  # Windows thumbnail cache
+        # IDE files
+        ".idea/",  # JetBrains IDEs
+        ".vscode/",  # VS Code
+        "*.sublime-*",  # Sublime Text
+        ".project",  # Eclipse
+        ".settings/",  # Eclipse
+        "*.code-workspace",  # VS Code workspace
+        # Environment files
+        ".env",  # Environment variables
+        ".venv/",  # Python virtual environments
+        "node_modules/",  # Node.js dependencies
+        "vendor/",  # Various dependencies
+        # Logs and caches
+        "*.log",  # Log files
+        ".cache/",  # Cache directories
+        ".pytest_cache/",  # Python test cache
+        "coverage/",  # Code coverage reports
+    ]  # Always ignore
     for path in gitignore_paths:
         if path.exists():
             with open(path) as f:
@@ -82,12 +64,12 @@ class FileWatcher:
     """Watches source files for changes and AI comments"""
 
     # Compiled regex pattern for AI comments
-    ai_comment_pattern = re.compile(r"(?:#|//|--) *(ai\b.*|ai\b.*|.*\bai!?) *$", re.IGNORECASE)
+    ai_comment_pattern = re.compile(r"(?:#|//|--) *(ai\b.*|ai\b.*|.*\bai[?!]?) *$", re.IGNORECASE)
 
-    def __init__(self, coder, gitignores=None, verbose=False, analytics=None):
+    def __init__(self, coder, gitignores=None, verbose=False, analytics=None, root=None):
         self.coder = coder
         self.io = coder.io
-        self.root = Path(coder.root)
+        self.root = Path(root) if root else Path(coder.root)
         self.verbose = verbose
         self.analytics = analytics
         self.stop_event = None
@@ -113,10 +95,9 @@ class FileWatcher:
         if self.verbose:
             dump(rel_path)
 
-        if self.gitignore_spec and self.gitignore_spec.match_file(str(rel_path)):
-            return False
-
-        if not is_source_file(path_obj):
+        if self.gitignore_spec and self.gitignore_spec.match_file(
+            rel_path.as_posix() + ("/" if path_abs.is_dir() else "")
+        ):
             return False
 
         if self.verbose:
@@ -129,28 +110,52 @@ class FileWatcher:
         except Exception:
             return
 
+    def get_roots_to_watch(self):
+        """Determine which root paths to watch based on gitignore rules"""
+        if self.gitignore_spec:
+            roots = [
+                str(path)
+                for path in self.root.iterdir()
+                if not self.gitignore_spec.match_file(
+                    path.relative_to(self.root).as_posix() + ("/" if path.is_dir() else "")
+                )
+            ]
+            # Fallback to watching root if all top-level items are filtered out
+            return roots if roots else [str(self.root)]
+        return [str(self.root)]
+
+    def handle_changes(self, changes):
+        """Process the detected changes and update state"""
+        if not changes:
+            return False
+
+        changed_files = {str(Path(change[1])) for change in changes}
+        self.changed_files.update(changed_files)
+        self.io.interrupt_input()
+        return True
+
+    def watch_files(self):
+        """Watch for file changes and process them"""
+        try:
+            roots_to_watch = self.get_roots_to_watch()
+
+            for changes in watch(
+                *roots_to_watch, watch_filter=self.filter_func, stop_event=self.stop_event
+            ):
+                if self.handle_changes(changes):
+                    return
+
+        except Exception as e:
+            if self.verbose:
+                dump(f"File watcher error: {e}")
+            raise e
+
     def start(self):
         """Start watching for file changes"""
         self.stop_event = threading.Event()
         self.changed_files = set()
 
-        def watch_files():
-            try:
-                for changes in watch(
-                    str(self.root), watch_filter=self.filter_func, stop_event=self.stop_event
-                ):
-                    if not changes:
-                        continue
-                    changed_files = {str(Path(change[1])) for change in changes}
-                    self.changed_files.update(changed_files)
-                    self.io.interrupt_input()
-                    return
-            except Exception as e:
-                if self.verbose:
-                    dump(f"File watcher error: {e}")
-                raise e
-
-        self.watcher_thread = threading.Thread(target=watch_files, daemon=True)
+        self.watcher_thread = threading.Thread(target=self.watch_files, daemon=True)
         self.watcher_thread.start()
 
     def stop(self):
@@ -165,10 +170,12 @@ class FileWatcher:
     def process_changes(self):
         """Get any detected file changes"""
 
-        has_bangs = False
+        has_action = None
+        added = False
         for fname in self.changed_files:
-            _, _, has_bang = self.get_ai_comments(fname)
-            has_bangs |= has_bang
+            _, _, action = self.get_ai_comments(fname)
+            if action in ("!", "?"):
+                has_action = action
 
             if fname in self.coder.abs_fnames:
                 continue
@@ -176,26 +183,30 @@ class FileWatcher:
                 self.analytics.event("ai-comments file-add")
             self.coder.abs_fnames.add(fname)
             rel_fname = self.coder.get_rel_fname(fname)
+            if not added:
+                self.io.tool_output()
+                added = True
             self.io.tool_output(f"Added {rel_fname} to the chat")
-            self.io.tool_output()
 
-        if not has_bangs:
+        if not has_action:
+            if added:
+                self.io.tool_output(
+                    "End your comment with AI! to request changes or AI? to ask questions"
+                )
             return ""
 
         if self.analytics:
             self.analytics.event("ai-comments execute")
         self.io.tool_output("Processing your request...")
 
-        res = """
-The "AI" comments below marked with █ can be found in the code files I've shared with you.
-They contain your instructions.
-Make the requested changes.
-Be sure to remove all these "AI" comments from the code!
-"""
+        if has_action == "!":
+            res = watch_code_prompt
+        elif has_action == "?":
+            res = watch_ask_prompt
 
         # Refresh all AI comments from tracked files
         for fname in self.coder.abs_fnames:
-            line_nums, comments, _has_bang = self.get_ai_comments(fname)
+            line_nums, comments, _action = self.get_ai_comments(fname)
             if not line_nums:
                 continue
 
@@ -233,11 +244,14 @@ Be sure to remove all these "AI" comments from the code!
         return res
 
     def get_ai_comments(self, filepath):
-        """Extract AI comment line numbers, comments and bang status from a file"""
+        """Extract AI comment line numbers, comments and action status from a file"""
         line_nums = []
         comments = []
-        has_bang = False
+        has_action = None  # None, "!" or "?"
         content = self.io.read_text(filepath, silent=True)
+        if not content:
+            return None, None, None
+
         for i, line in enumerate(content.splitlines(), 1):
             if match := self.ai_comment_pattern.search(line):
                 comment = match.group(0).strip()
@@ -248,10 +262,12 @@ Be sure to remove all these "AI" comments from the code!
                     comment = comment.lstrip("/#-")
                     comment = comment.strip()
                     if comment.startswith("ai!") or comment.endswith("ai!"):
-                        has_bang = True
+                        has_action = "!"
+                    elif comment.startswith("ai?") or comment.endswith("ai?"):
+                        has_action = "?"
         if not line_nums:
-            return None, None, False
-        return line_nums, comments, has_bang
+            return None, None, None
+        return line_nums, comments, has_action
 
 
 def main():

aider/watch_prompts.py

@@ -0,0 +1,12 @@
+watch_code_prompt = """
+I've written your instructions in comments in the code and marked them with "ai"
+You can see the "AI" comments shown below (marked with █).
+Find them in the code files I've shared with you, and follow their instructions.
+
+After completing those instructions, also be sure to remove all the "AI" comments from the code too.
+"""
+
+watch_ask_prompt = """/ask
+Find the "AI" comments below (marked with █) in the code files I've shared with you.
+They contain my questions that I need you to answer and other instructions for you.
+"""

aider/website/HISTORY.md

@@ -1,7 +1,6 @@
 ---
 title: Release history
-parent: More info
-nav_order: 900
+nav_order: 925
 highlight_image: /assets/blame.jpg
 description: Release notes and stats on aider writing its own code.
 ---
@@ -24,7 +23,125 @@ cog.out(text)
 ]]]-->
 
 
-### main branch
+### Aider v0.74.0
+
+- Dynamically changes the Ollama context window to hold the current chat.
+- Better support for o3-mini, DeepSeek V3 & R1, o1-mini, o1 via secondary API providers.
+- Remove `<think>` tags from R1 responses for commit messages (and other weak model uses).
+- Can now specify `use_temperature: <float>` in model settings, not just true/false.
+- The full docker container now includes `boto3` for Bedrock.
+- Docker containers now set `HOME=/app` which is the normal project mount-point, to persist `~/.aider`.
+- Bugfix to prevent creating incorrect filenames like `python`, `php`, etc.
+- Bugfix for `--timeout`
+- Bugfix so that `/model` now correctly reports that the weak model is not changed.
+- Bugfix so that multi-line mode persists through ^C at confirmation prompts.
+- Watch files now fully ignores top-level directories named in ignore files, to reduce the chance of hitting OS watch limits. Helpful to ignore giant subtrees like `node_modules`.
+- Fast startup with more providers and when model metadata provided in local files.
+- Improved .gitignore handling:
+  - Honor ignores already in effect regardless of how they've been configured.
+  - Check for .env only when the file exists.
+- Yes/No prompts now accept All/Skip as alias for Y/N even when not processing a group of confirmations.
+- Aider wrote 77% of the code in this release.
+
+### Aider v0.73.0
+
+- Full support for o3-mini: `aider --model o3-mini`
+- New `--reasoning-effort` argument: low, medium, high.
+- Improved handling of context window size limits, with better messaging and Ollama-specific guidance.
+- Added support for removing model-specific reasoning tags from responses with `remove_reasoning: tagname` model setting.
+- Auto-create parent directories when creating new files, by xqyz.
+- Support for R1 free on OpenRouter: `--model openrouter/deepseek/deepseek-r1:free`
+- Aider wrote 69% of the code in this release.
+
+### Aider v0.72.3
+
+- Enforce user/assistant turn order to avoid R1 errors, by miradnanali.
+- Case-insensitive model name matching while preserving original case.
+
+### Aider v0.72.2
+- Harden against user/assistant turn order problems which cause R1 errors.
+
+### Aider v0.72.1
+- Fix model metadata for `openrouter/deepseek/deepseek-r1`
+
+### Aider v0.72.0
+- Support for DeepSeek R1.
+  - Use shortcut: `--model r1`
+  - Also via OpenRouter: `--model openrouter/deepseek/deepseek-r1`
+- Added Kotlin syntax support to repo map, by Paul Walker.
+- Added `--line-endings` for file writing, by Titusz Pan.
+- Added examples_as_sys_msg=True for GPT-4o models, improves benchmark scores.
+- Bumped all dependencies, to pick up litellm support for o1 system messages.
+- Bugfix for turn taking when reflecting lint/test errors.
+- Aider wrote 52% of the code in this release.
+
+### Aider v0.71.1
+
+- Fix permissions issue in Docker images.
+- Added read-only file announcements.
+- Bugfix: ASCII fallback for unicode errors.
+- Bugfix: integer indices for list slicing in repomap calculations.
+
+### Aider v0.71.0
+
+- Prompts to help DeepSeek work better when alternating between `/ask` and `/code`.
+- Streaming pretty LLM responses is smoother and faster for long replies.
+- Streaming automatically turns of for model that don't support it
+  - Can now switch to/from `/model o1` and a streaming model
+- Pretty output remains enabled even when editing files with triple-backtick fences
+- Bare `/ask`, `/code` and `/architect` commands now switch the chat mode.
+- Increased default size of the repomap.
+- Increased max chat history tokens limit from 4k to 8k.
+- Turn off fancy input and watch files if terminal is dumb.
+- Added support for custom voice format and input device settings.
+- Disabled Streamlit email prompt, by apaz-cli.
+- Docker container runs as non-root user.
+- Fixed lint command handling of nested spaced strings, by Aaron Weisberg.
+- Added token count feedback when adding command output to chat.
+- Improved error handling for large audio files with automatic format conversion.
+- Improved handling of git repo index errors, by Krazer.
+- Improved unicode handling in console output with ASCII fallback.
+- Added AssertionError, AttributeError to git error handling.
+- Aider wrote 60% of the code in this release.
+
+### Aider v0.70.0
+
+- Full support for o1 models.
+- Watch files now honors `--subtree-only`, and only watches that subtree.
+- Improved prompting for watch files, to work more reliably with more models.
+- New install methods via uv, including one-liners.
+- Support for openrouter/deepseek/deepseek-chat model.
+- Better error handling when interactive commands are attempted via `/load` or `--load`.
+- Display read-only files with abs path if its shorter than rel path.
+- Ask 10% of users to opt-in to analytics.
+- Bugfix for auto-suggest.
+- Gracefully handle unicode errors in git path names.
+- Aider wrote 74% of the code in this release.
+
+### Aider v0.69.1
+
+- Fix for gemini model names in model metadata.
+- Show hints about AI! and AI? when user makes AI comments.
+- Support for running without git installed.
+- Improved environment variable setup messages on Windows.
+
+### Aider v0.69.0
+
+- [Watch files](https://aider.chat/docs/usage/watch.html) improvements:
+  - Use `# ... AI?` comments to trigger aider and ask questions about your code.
+  - Now watches *all* files, not just certain source files.
+  - Use `# AI comments`, `// AI comments`, or `-- AI comments` to give aider instructions in any text file.
+- Full support for Gemini Flash 2.0 Exp:
+  - `aider --model flash` or `aider --model gemini/gemini-2.0-flash-exp`
+- [New `--multiline` flag and `/multiline-mode` command](https://aider.chat/docs/usage/commands.html#entering-multi-line-chat-messages) makes ENTER a soft newline and META-ENTER send the message, by @miradnanali.
+- `/copy-context <instructions>` now takes optional "instructions" when [copying code context to the clipboard](https://aider.chat/docs/usage/copypaste.html#copy-aiders-code-context-to-your-clipboard-paste-into-the-web-ui).
+- Improved clipboard error handling with helpful requirements install info.
+- Ask 5% of users if they want to opt-in to analytics.
+- `/voice` now lets you edit the transcribed text before sending.
+- Disabled auto-complete in Y/N prompts.
+- Aider wrote 68% of the code in this release.
+
+### Aider v0.68.0
 
 - [Aider works with LLM web chat UIs](https://aider.chat/docs/usage/copypaste.html).
   - New `--copy-paste` mode.

aider/website/_config.yml

@@ -1,6 +1,12 @@
 theme: just-the-docs
 url: "https://aider.chat"
 
+# Analytics configuration
+analytics:
+  enabled: false  # Single switch to control analytics and cookie consent
+  posthog_key: 'phc_99T7muzafUMMZX15H8XePbMSreEUzahHbtWjy3l5Qbv'
+  posthog_host: 'https://us.i.posthog.com'
+
 plugins:
   - jekyll-redirect-from
   - jekyll-sitemap
@@ -45,4 +51,4 @@ callouts:
   note:
     title: Note
     color: yellow
-    
+    

aider/website/_data/blame.yml

@@ -3252,8 +3252,8 @@
     Philippe de Reynal: 30
   start_tag: v0.65.0
   total_lines: 703
-- aider_percentage: 65.29
-  aider_total: 457
+- aider_percentage: 67.86
+  aider_total: 437
   end_date: '2024-12-06'
   end_tag: v0.67.0
   file_counts:
@@ -3314,14 +3314,428 @@
     tests/browser/test_browser.py:
       Paul Gauthier: 2
       Paul Gauthier (aider): 1
-    tests/fixtures/watch.js:
-      Paul Gauthier: 19
-      Paul Gauthier (aider): 16
-    tests/fixtures/watch.py:
-      Paul Gauthier: 17
-      Paul Gauthier (aider): 4
   grand_total:
-    Paul Gauthier: 243
-    Paul Gauthier (aider): 457
+    Paul Gauthier: 207
+    Paul Gauthier (aider): 437
   start_tag: v0.66.0
-  total_lines: 700
+  total_lines: 644
+- aider_percentage: 71.57
+  aider_total: 428
+  end_date: '2024-12-10'
+  end_tag: v0.68.0
+  file_counts:
+    .github/workflows/pages.yml:
+      Paul Gauthier (aider): 1
+    aider/__init__.py:
+      Paul Gauthier: 1
+    aider/analytics.py:
+      Paul Gauthier: 2
+    aider/args.py:
+      Paul Gauthier: 30
+      Paul Gauthier (aider): 15
+    aider/coders/base_coder.py:
+      Paul Gauthier: 12
+      Paul Gauthier (aider): 10
+    aider/coders/editor_editblock_coder.py:
+      Paul Gauthier: 1
+    aider/coders/editor_whole_coder.py:
+      Paul Gauthier: 1
+    aider/commands.py:
+      Paul Gauthier: 24
+      Paul Gauthier (aider): 25
+    aider/copypaste.py:
+      Paul Gauthier: 4
+      Paul Gauthier (aider): 68
+    aider/exceptions.py:
+      Paul Gauthier: 9
+    aider/history.py:
+      Paul Gauthier (aider): 1
+    aider/io.py:
+      Paul Gauthier: 13
+    aider/main.py:
+      Paul Gauthier: 17
+      Paul Gauthier (aider): 29
+    aider/models.py:
+      Paul Gauthier: 15
+    aider/repo.py:
+      Paul Gauthier (aider): 1
+    aider/run_cmd.py:
+      Paul Gauthier: 1
+    aider/sendchat.py:
+      Paul Gauthier (aider): 4
+    aider/utils.py:
+      Paul Gauthier: 1
+    aider/voice.py:
+      Paul Gauthier: 9
+    aider/watch.py:
+      Paul Gauthier: 6
+      Paul Gauthier (aider): 35
+    aider/website/_includes/edit-leaderboard.js:
+      Paul Gauthier: 2
+      Paul Gauthier (aider): 90
+    aider/website/_includes/head_custom.html:
+      Paul Gauthier: 7
+      Paul Gauthier (aider): 53
+    aider/website/docs/leaderboards/index.md:
+      Paul Gauthier: 1
+      Paul Gauthier (aider): 2
+    scripts/blame.py:
+      Paul Gauthier: 6
+      Paul Gauthier (aider): 17
+    scripts/pip-compile.sh:
+      Paul Gauthier: 2
+    tests/basic/test_commands.py:
+      Paul Gauthier (aider): 24
+    tests/basic/test_history.py:
+      Paul Gauthier (aider): 3
+    tests/basic/test_main.py:
+      Paul Gauthier (aider): 46
+    tests/basic/test_repo.py:
+      Paul Gauthier (aider): 3
+    tests/basic/test_sendchat.py:
+      Paul Gauthier: 5
+      Paul Gauthier (aider): 1
+    tests/basic/test_watch.py:
+      Paul Gauthier: 1
+  grand_total:
+    Paul Gauthier: 170
+    Paul Gauthier (aider): 428
+  start_tag: v0.67.0
+  total_lines: 598
+- aider_percentage: 67.87
+  aider_total: 207
+  end_date: '2024-12-13'
+  end_tag: v0.69.0
+  file_counts:
+    .github/workflows/pages.yml:
+      Paul Gauthier: 2
+    aider/__init__.py:
+      Paul Gauthier: 1
+    aider/analytics.py:
+      Paul Gauthier: 2
+    aider/args.py:
+      Mir Adnan ALI: 3
+      Paul Gauthier: 1
+    aider/coders/base_coder.py:
+      JeongJuhyeon: 1
+      Mir Adnan ALI: 3
+    aider/commands.py:
+      Mir Adnan ALI: 4
+      Paul Gauthier: 5
+      Paul Gauthier (aider): 3
+    aider/io.py:
+      Mir Adnan ALI: 37
+      Paul Gauthier: 8
+      Paul Gauthier (aider): 3
+    aider/main.py:
+      Mir Adnan ALI: 1
+    aider/models.py:
+      Paul Gauthier: 7
+    aider/watch.py:
+      Paul Gauthier: 7
+      Paul Gauthier (aider): 47
+    aider/website/docs/leaderboards/index.md:
+      Paul Gauthier: 1
+    benchmark/benchmark.py:
+      Paul Gauthier: 7
+      Paul Gauthier (aider): 7
+    scripts/blame.py:
+      Paul Gauthier (aider): 1
+    scripts/issues.py:
+      Paul Gauthier (aider): 58
+    scripts/update-history.py:
+      Paul Gauthier: 3
+    tests/basic/test_io.py:
+      Paul Gauthier (aider): 20
+    tests/basic/test_watch.py:
+      Paul Gauthier: 5
+      Paul Gauthier (aider): 68
+  grand_total:
+    JeongJuhyeon: 1
+    Mir Adnan ALI: 48
+    Paul Gauthier: 49
+    Paul Gauthier (aider): 207
+  start_tag: v0.68.0
+  total_lines: 305
+- aider_percentage: 74.22
+  aider_total: 875
+  end_date: '2024-12-26'
+  end_tag: v0.70.0
+  file_counts:
+    aider/__init__.py:
+      Paul Gauthier: 1
+    aider/analytics.py:
+      Paul Gauthier: 6
+      Paul Gauthier (aider): 41
+    aider/args.py:
+      Evan Johnson: 2
+    aider/coders/search_replace.py:
+      Paul Gauthier: 5
+    aider/commands.py:
+      Paul Gauthier (aider): 41
+    aider/help_pats.py:
+      Paul Gauthier: 3
+    aider/io.py:
+      Paul Gauthier: 7
+      Paul Gauthier (aider): 9
+    aider/main.py:
+      Paul Gauthier: 15
+      Paul Gauthier (aider): 5
+      apaz-cli: 3
+      mdk: 6
+    aider/models.py:
+      Paul Gauthier: 29
+    aider/repo.py:
+      Paul Gauthier: 14
+    aider/utils.py:
+      Paul Gauthier: 2
+    aider/watch.py:
+      Paul Gauthier: 13
+    aider/website/_includes/head_custom.html:
+      Paul Gauthier (aider): 4
+    aider/website/_includes/leaderboard.js:
+      Paul Gauthier (aider): 14
+    aider/website/docs/leaderboards/index.md:
+      Paul Gauthier: 28
+      Paul Gauthier (aider): 2
+    benchmark/Dockerfile:
+      Paul Gauthier: 8
+      Paul Gauthier (aider): 43
+    benchmark/benchmark.py:
+      Paul Gauthier: 69
+      Paul Gauthier (aider): 153
+    benchmark/clone-exercism.sh:
+      Paul Gauthier: 2
+      Paul Gauthier (aider): 18
+    benchmark/cpp-test.sh:
+      Paul Gauthier: 10
+      Paul Gauthier (aider): 1
+    benchmark/docker.sh:
+      Paul Gauthier (aider): 4
+    benchmark/install-docker-ubuntu.sh:
+      Paul Gauthier (aider): 63
+    benchmark/npm-test.sh:
+      Paul Gauthier: 10
+      Paul Gauthier (aider): 3
+    benchmark/problem_stats.py:
+      Paul Gauthier: 35
+      Paul Gauthier (aider): 318
+    benchmark/rsync.sh:
+      Paul Gauthier: 7
+      Paul Gauthier (aider): 26
+    scripts/blame.py:
+      Paul Gauthier (aider): 6
+    scripts/my_models.py:
+      Paul Gauthier (aider): 95
+    scripts/update-blame.sh:
+      Paul Gauthier (aider): 3
+    scripts/update-docs.sh:
+      Paul Gauthier: 1
+    tests/basic/test_analytics.py:
+      Paul Gauthier (aider): 19
+    tests/basic/test_main.py:
+      Paul Gauthier (aider): 7
+    tests/basic/test_sanity_check_repo.py:
+      mdk: 28
+  grand_total:
+    Evan Johnson: 2
+    Paul Gauthier: 265
+    Paul Gauthier (aider): 875
+    apaz-cli: 3
+    mdk: 34
+  start_tag: v0.69.0
+  total_lines: 1179
+- aider_percentage: 60.36
+  aider_total: 236
+  end_date: '2025-01-10'
+  end_tag: v0.71.0
+  file_counts:
+    aider/__init__.py:
+      Paul Gauthier: 1
+    aider/args.py:
+      Paul Gauthier: 2
+    aider/coders/base_coder.py:
+      Paul Gauthier: 7
+      Paul Gauthier (aider): 13
+    aider/commands.py:
+      Paul Gauthier: 1
+      Paul Gauthier (aider): 22
+    aider/io.py:
+      Paul Gauthier: 3
+      Paul Gauthier (aider): 16
+    aider/linter.py:
+      Aaron Weisberg: 5
+    aider/main.py:
+      Paul Gauthier: 7
+      Paul Gauthier (aider): 13
+      apaz-cli: 18
+    aider/mdstream.py:
+      Paul Gauthier: 38
+      Paul Gauthier (aider): 58
+    aider/models.py:
+      Paul Gauthier: 11
+      Paul Gauthier (aider): 2
+    aider/repo.py:
+      Krazer: 10
+      Paul Gauthier: 5
+    aider/run_cmd.py:
+      Aaron Weisberg: 2
+    aider/utils.py:
+      Paul Gauthier: 9
+    aider/voice.py:
+      Paul Gauthier: 11
+      Paul Gauthier (aider): 13
+    aider/watch.py:
+      Paul Gauthier: 1
+    benchmark/Dockerfile:
+      Josh Vera: 1
+      Paul Maunders: 12
+    benchmark/benchmark.py:
+      Nimesh Ghelani: 1
+      Paul Gauthier: 6
+      Paul Gauthier (aider): 30
+    benchmark/problem_stats.py:
+      Paul Gauthier (aider): 5
+    docker/Dockerfile:
+      Paul Gauthier (aider): 32
+    scripts/update-history.py:
+      Paul Gauthier (aider): 1
+    tests/basic/test_commands.py:
+      Paul Gauthier: 2
+    tests/basic/test_io.py:
+      Paul Gauthier (aider): 6
+    tests/basic/test_linter.py:
+      Aaron Weisberg: 2
+    tests/basic/test_models.py:
+      Paul Gauthier (aider): 25
+  grand_total:
+    Aaron Weisberg: 9
+    Josh Vera: 1
+    Krazer: 10
+    Nimesh Ghelani: 1
+    Paul Gauthier: 104
+    Paul Gauthier (aider): 236
+    Paul Maunders: 12
+    apaz-cli: 18
+  start_tag: v0.70.0
+  total_lines: 391
+- aider_percentage: 51.69
+  aider_total: 138
+  end_date: '2025-01-20'
+  end_tag: v0.72.0
+  file_counts:
+    .github/workflows/docker-build-test.yml:
+      Paul Gauthier (aider): 38
+    .github/workflows/pages.yml:
+      Paul Gauthier: 3
+      Paul Gauthier (aider): 1
+    .github/workflows/ubuntu-tests.yml:
+      Paul Gauthier (aider): 8
+    .github/workflows/windows-tests.yml:
+      Paul Gauthier (aider): 8
+    aider/__init__.py:
+      Paul Gauthier: 1
+    aider/args.py:
+      Titusz Pan: 6
+    aider/coders/base_coder.py:
+      Paul Gauthier: 11
+    aider/coders/single_wholefile_func_coder.py:
+      Paul Gauthier: 1
+    aider/coders/wholefile_func_coder.py:
+      Paul Gauthier: 1
+    aider/commands.py:
+      Paul Gauthier: 3
+    aider/history.py:
+      Paul Gauthier: 7
+    aider/io.py:
+      Paul Gauthier (aider): 14
+      Titusz Pan: 2
+    aider/main.py:
+      Titusz Pan: 1
+    aider/models.py:
+      Paul Gauthier: 16
+    aider/queries/tree-sitter-kotlin-tags.scm:
+      Paul Walker: 27
+    aider/repomap.py:
+      Paul Gauthier (aider): 2
+    aider/sendchat.py:
+      Paul Gauthier: 9
+      Paul Gauthier (aider): 22
+    aider/utils.py:
+      Paul Gauthier: 1
+    aider/website/docs/leaderboards/index.md:
+      Paul Gauthier: 2
+    benchmark/benchmark.py:
+      Paul Gauthier: 9
+    benchmark/rsync.sh:
+      Paul Gauthier: 21
+    docker/Dockerfile:
+      Paul Gauthier: 2
+      Paul Gauthier (aider): 6
+    scripts/my_models.py:
+      Paul Gauthier: 3
+    scripts/update-docs.sh:
+      Paul Gauthier: 2
+    tests/basic/test_io.py:
+      Paul Gauthier (aider): 39
+    tests/basic/test_repomap.py:
+      Paul Walker: 1
+  grand_total:
+    Paul Gauthier: 92
+    Paul Gauthier (aider): 138
+    Paul Walker: 28
+    Titusz Pan: 9
+  start_tag: v0.71.0
+  total_lines: 267
+- aider_percentage: 69.44
+  aider_total: 284
+  end_date: '2025-01-31'
+  end_tag: v0.73.0
+  file_counts:
+    aider/__init__.py:
+      Paul Gauthier: 1
+    aider/args.py:
+      Paul Gauthier: 3
+      Paul Gauthier (aider): 2
+    aider/coders/base_coder.py:
+      Paul Gauthier: 37
+      Paul Gauthier (aider): 26
+    aider/commands.py:
+      xqyz: 1
+    aider/io.py:
+      Paul Gauthier: 7
+    aider/main.py:
+      Paul Gauthier: 13
+      Paul Gauthier (aider): 15
+    aider/models.py:
+      Paul Gauthier: 8
+      Paul Gauthier (aider): 33
+    aider/sendchat.py:
+      Mir Adnan ALI: 28
+      Paul Gauthier: 11
+      Paul Gauthier (aider): 6
+    aider/urls.py:
+      Paul Gauthier: 1
+    aider/website/_includes/leaderboard.js:
+      Paul Gauthier (aider): 1
+    aider/website/docs/leaderboards/index.md:
+      Paul Gauthier: 3
+      Paul Gauthier (aider): 2
+    benchmark/benchmark.py:
+      Paul Gauthier (aider): 21
+    benchmark/rsync.sh:
+      Paul Gauthier: 2
+    tests/basic/test_coder.py:
+      Paul Gauthier: 10
+      Paul Gauthier (aider): 39
+    tests/basic/test_main.py:
+      Paul Gauthier (aider): 62
+    tests/basic/test_sendchat.py:
+      Paul Gauthier (aider): 77
+  grand_total:
+    Mir Adnan ALI: 28
+    Paul Gauthier: 96
+    Paul Gauthier (aider): 284
+    xqyz: 1
+  start_tag: v0.72.0
+  total_lines: 409

aider/website/_data/deepseek-down.yml

@@ -0,0 +1,130 @@
+- dirname: 2024-12-25-13-31-51--deepseekv3preview-diff2
+  test_cases: 225
+  model: DeepSeek
+  edit_format: diff
+  commit_hash: 0a23c4a-dirty
+  pass_rate_1: 22.7
+  pass_rate_2: 48.4
+  pass_num_1: 51
+  pass_num_2: 109
+  percent_cases_well_formed: 98.7
+  error_outputs: 7
+  num_malformed_responses: 7
+  num_with_malformed_responses: 3
+  user_asks: 19
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 0
+  test_timeouts: 8
+  total_tests: 225
+  command: aider --model deepseek/deepseek-chat
+  date: 2024-12-25
+  versions: 0.69.2.dev
+  seconds_per_case: 34.8
+  total_cost: 0.3369
+
+
+- dirname: 2025-01-28-17-47-49--v3-fireworks
+  test_cases: 225
+  model: Fireworks
+  edit_format: diff
+  commit_hash: 0336a98-dirty
+  pass_rate_1: 22.2
+  pass_rate_2: 48.4
+  pass_num_1: 50
+  pass_num_2: 109
+  percent_cases_well_formed: 96.9
+  error_outputs: 18
+  num_malformed_responses: 16
+  num_with_malformed_responses: 7
+  user_asks: 14
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 2
+  test_timeouts: 9
+  total_tests: 225
+  command: aider --model fireworks_ai/accounts/fireworks/models/deepseek-v3
+  date: 2025-01-28
+  versions: 0.72.4.dev
+  seconds_per_case: 115.9
+  total_cost: 2.1177
+
+- dirname: 2025-01-28-19-25-32--or-v3-deepinfra-diff
+  test_cases: 222
+  model: "OpenRouter: DeepInfra"
+  edit_format: diff
+  commit_hash: bfc5745, 77d2bc5-dirty
+  pass_rate_1: 23.9
+  pass_rate_2: 48.0
+  pass_num_1: 53
+  pass_num_2: 108
+  percent_cases_well_formed: 99.5
+  error_outputs: 18
+  num_malformed_responses: 1
+  num_with_malformed_responses: 1
+  user_asks: 17
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 2
+  test_timeouts: 4
+  total_tests: 225
+  command: aider --model openrouter/deepseek/deepseek-chat
+  date: 2025-01-28
+  versions: 0.72.4.dev
+  seconds_per_case: 187.0
+  total_cost: 0.2733
+
+- dirname: 2025-01-28-21-07-23--or-v3-novita-diff
+  test_cases: 225
+  model: "OpenRouter: Novita"
+  edit_format: diff
+  commit_hash: 66025a0
+  pass_rate_1: 20.4
+  pass_rate_2: 42.7
+  pass_num_1: 46
+  pass_num_2: 96
+  percent_cases_well_formed: 84.0
+  error_outputs: 265
+  num_malformed_responses: 67
+  num_with_malformed_responses: 36
+  user_asks: 5
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 0
+  test_timeouts: 8
+  total_tests: 225
+  command: aider --model openrouter/deepseek/deepseek-chat
+  date: 2025-01-28
+  versions: 0.72.4.dev
+  seconds_per_case: 472.5
+  total_cost: 0.0000
+
+- dirname: 2025-01-29-00-36-49--v3-hyperolic-diff
+  test_cases: 224
+  model: Hyperbolic
+  edit_format: diff
+  commit_hash: 298f713
+  pass_rate_1: 20.5
+  pass_rate_2: 48.4
+  pass_num_1: 46
+  pass_num_2: 109
+  percent_cases_well_formed: 97.3
+  error_outputs: 29
+  num_malformed_responses: 6
+  num_with_malformed_responses: 6
+  user_asks: 7
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 0
+  test_timeouts: 7
+  total_tests: 225
+  command: OPENAI_API_BASE=https://api.hyperbolic.xyz/v1/ aider --model openai/deepseek-ai/DeepSeek-V3
+  date: 2025-01-29
+  versions: 0.72.4.dev
+  seconds_per_case: 365.4
+  total_cost: 0.0000

aider/website/_data/edit_leaderboard.yml

@@ -1232,7 +1232,7 @@
 
 - dirname: 2024-09-24-16-33-23--gemini-1.5-flash-002-whole
   test_cases: 133
-  model: gemini-1.5-flash-002
+  model: gemini-1.5-flash-002 (0924)
   edit_format: whole
   commit_hash: 3edcd71
   pass_rate_1: 37.6
@@ -2134,4 +2134,99 @@
   date: 2024-12-10
   versions: 0.67.1.dev
   seconds_per_case: 32.7
-  total_cost: 0.1106
+  total_cost: 0.1106
+
+- dirname: 2024-12-11-00-37-08--yi-test
+  test_cases: 133
+  model: yi-lightning
+  edit_format: whole
+  commit_hash: e909a3d-dirty
+  pass_rate_1: 49.6
+  pass_rate_2: 65.4
+  percent_cases_well_formed: 97.0
+  error_outputs: 304
+  num_malformed_responses: 5
+  num_with_malformed_responses: 4
+  user_asks: 34
+  lazy_comments: 2
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 0
+  test_timeouts: 2
+  command: aider --model openai/yi-lightning
+  date: 2024-12-11
+  versions: 0.67.1.dev
+  seconds_per_case: 57.8
+  total_cost: 0.0000
+
+- dirname: 2024-12-11-21-07-00--gemini-2-flash-diff
+  test_cases: 133
+  model: gemini-2.0-flash-exp
+  edit_format: diff
+  commit_hash: fcb2bac-dirty, 02e7e31-dirty
+  pass_rate_1: 56.4
+  pass_rate_2: 69.9
+  percent_cases_well_formed: 97.0
+  error_outputs: 10
+  num_malformed_responses: 6
+  num_with_malformed_responses: 4
+  user_asks: 8
+  lazy_comments: 0
+  syntax_errors: 1
+  indentation_errors: 0
+  exhausted_context_windows: 2
+  test_timeouts: 1
+  command: aider --model gemini/gemini-2.0-flash-exp
+  date: 2024-12-11
+  versions: 0.68.1.dev
+  seconds_per_case: 7.3
+  total_cost: 0.0000
+
+- dirname: 2024-12-18-01-50-08--o1
+  test_cases: 133
+  model: o1
+  edit_format: diff
+  commit_hash: 074c636-dirty
+  pass_rate_1: 65.4
+  pass_rate_2: 84.2
+  percent_cases_well_formed: 99.2
+  error_outputs: 1
+  num_malformed_responses: 1
+  num_with_malformed_responses: 1
+  user_asks: 0
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 0
+  test_timeouts: 2
+  command: aider --model openrouter/openai/o1
+  date: 2024-12-18
+  versions: 0.69.2.dev
+  seconds_per_case: 29.9
+  total_cost: 0.0000
+
+- dirname: 2024-12-21-22-06-01--polyglot-o1-mini-whole
+  test_cases: 225
+  model: o1-mini-2024-09-12
+  edit_format: whole
+  commit_hash: a755079-dirty
+  pass_rate_1: 8.9
+  pass_rate_2: 27.1
+  pass_num_1: 20
+  pass_num_2: 61
+  percent_cases_well_formed: 95.6
+  error_outputs: 15
+  num_malformed_responses: 14
+  num_with_malformed_responses: 10
+  user_asks: 37
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 0
+  test_timeouts: 5
+  total_tests: 225
+  command: aider --model o1-mini
+  date: 2024-12-21
+  versions: 0.69.2.dev
+  seconds_per_case: 34.3
+  total_cost: 17.6270

aider/website/_data/o1_polyglot_leaderboard.yml

@@ -0,0 +1,259 @@
+- dirname: 2024-12-21-18-41-18--polyglot-gpt-4o-mini
+  test_cases: 225
+  model: gpt-4o-mini-2024-07-18
+  edit_format: whole
+  commit_hash: a755079-dirty
+  pass_rate_1: 0.9
+  pass_rate_2: 3.6
+  pass_num_1: 2
+  pass_num_2: 8
+  percent_cases_well_formed: 100.0
+  error_outputs: 0
+  num_malformed_responses: 0
+  num_with_malformed_responses: 0
+  user_asks: 36
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 0
+  test_timeouts: 3
+  total_tests: 225
+  command: aider --model gpt-4o-mini-2024-07-18
+  date: 2024-12-21
+  versions: 0.69.2.dev
+  seconds_per_case: 17.3
+  total_cost: 0.3236
+
+- dirname: 2024-12-21-18-44-28--polyglot-sonnet
+  test_cases: 225
+  model: claude-3-5-sonnet-20241022
+  edit_format: diff
+  commit_hash: a755079-dirty
+  pass_rate_1: 18.7
+  pass_rate_2: 45.3
+  pass_num_1: 42
+  pass_num_2: 102
+  percent_cases_well_formed: 100.0
+  error_outputs: 1
+  num_malformed_responses: 0
+  num_with_malformed_responses: 0
+  user_asks: 14
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 1
+  test_timeouts: 12
+  total_tests: 225
+  command: aider --model claude-3-5-sonnet-20241022
+  date: 2024-12-21
+  versions: 0.69.2.dev
+  seconds_per_case: 30.8
+  total_cost: 13.4847
+
+- dirname: 2024-12-21-18-52-34--polyglot-gpt-4o-diff
+  test_cases: 225
+  model: gpt-4o-2024-11-20
+  edit_format: diff
+  commit_hash: a755079-dirty
+  pass_rate_1: 4.9
+  pass_rate_2: 15.1
+  pass_num_1: 11
+  pass_num_2: 34
+  percent_cases_well_formed: 96.0
+  error_outputs: 12
+  num_malformed_responses: 11
+  num_with_malformed_responses: 9
+  user_asks: 34
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 1
+  test_timeouts: 19
+  total_tests: 225
+  command: aider --model gpt-4o-2024-11-20
+  date: 2024-12-21
+  versions: 0.69.2.dev
+  seconds_per_case: 22.2
+  total_cost: 7.1835
+
+- dirname: 2024-12-21-19-23-03--polyglot-o1-hard-diff
+  test_cases: 224
+  model: o1-2024-12-17 (high)
+  edit_format: diff
+  commit_hash: a755079-dirty
+  pass_rate_1: 23.7
+  pass_rate_2: 61.7
+  pass_num_1: 53
+  pass_num_2: 139
+  percent_cases_well_formed: 91.5
+  error_outputs: 25
+  num_malformed_responses: 24
+  num_with_malformed_responses: 19
+  user_asks: 16
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 0
+  test_timeouts: 2
+  total_tests: 225
+  command: aider --model openrouter/openai/o1
+  date: 2024-12-21
+  versions: 0.69.2.dev
+  seconds_per_case: 133.2
+  total_cost: 0.0000
+
+- dirname: 2024-12-21-20-56-21--polyglot-deepseek-diff
+  test_cases: 225
+  model: DeepSeek Chat V2.5
+  edit_format: diff
+  commit_hash: a755079-dirty
+  pass_rate_1: 5.3
+  pass_rate_2: 17.8
+  pass_num_1: 12
+  pass_num_2: 40
+  percent_cases_well_formed: 92.9
+  error_outputs: 42
+  num_malformed_responses: 37
+  num_with_malformed_responses: 16
+  user_asks: 23
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 5
+  test_timeouts: 5
+  total_tests: 225
+  command: aider --model deepseek/deepseek-chat
+  date: 2024-12-21
+  versions: 0.69.2.dev
+  seconds_per_case: 184.0
+  total_cost: 0.5101
+
+- dirname: 2024-12-21-21-46-27--polyglot-haiku-diff
+  test_cases: 225
+  model: claude-3-5-haiku-20241022
+  edit_format: diff
+  commit_hash: a755079-dirty
+  pass_rate_1: 7.1
+  pass_rate_2: 28.0
+  pass_num_1: 16
+  pass_num_2: 63
+  percent_cases_well_formed: 91.1
+  error_outputs: 31
+  num_malformed_responses: 30
+  num_with_malformed_responses: 20
+  user_asks: 13
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 1
+  test_timeouts: 9
+  total_tests: 225
+  command: aider --model claude-3-5-haiku-20241022
+  date: 2024-12-21
+  versions: 0.69.2.dev
+  seconds_per_case: 31.8
+  total_cost: 6.0583
+
+- dirname: 2024-12-22-13-22-32--polyglot-qwen-diff
+  test_cases: 225
+  model: Qwen2.5-Coder-32B-Instruct
+  edit_format: diff
+  commit_hash: 6d7e8be-dirty
+  pass_rate_1: 4.4
+  pass_rate_2: 8.0
+  pass_num_1: 10
+  pass_num_2: 18
+  percent_cases_well_formed: 71.6
+  error_outputs: 158
+  num_malformed_responses: 148
+  num_with_malformed_responses: 64
+  user_asks: 132
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 1
+  test_timeouts: 2
+  total_tests: 225
+  command: "aider --model openai/Qwen/Qwen2.5-Coder-32B-Instruct # via hyperbolic"
+  date: 2024-12-22
+  versions: 0.69.2.dev
+  seconds_per_case: 84.4
+  total_cost: 0.0000
+
+- dirname: 2024-12-22-21-26-35--polyglot-o1mini-whole
+  test_cases: 225
+  model: o1-mini-2024-09-12
+  edit_format: whole
+  commit_hash: 37df899
+  pass_rate_1: 5.8
+  pass_rate_2: 32.9
+  pass_num_1: 13
+  pass_num_2: 74
+  percent_cases_well_formed: 96.9
+  error_outputs: 8
+  num_malformed_responses: 8
+  num_with_malformed_responses: 7
+  user_asks: 27
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 0
+  test_timeouts: 3
+  total_tests: 225
+  command: aider --model o1-mini
+  date: 2024-12-22
+  versions: 0.69.2.dev
+  seconds_per_case: 34.7
+  total_cost: 18.5770
+
+- dirname: 2024-12-22-18-43-25--gemini-exp-1206-polyglot-whole-2
+  test_cases: 225
+  model: gemini-exp-1206
+  edit_format: whole
+  commit_hash: b1bc2f8
+  pass_rate_1: 19.6
+  pass_rate_2: 38.2
+  pass_num_1: 44
+  pass_num_2: 86
+  percent_cases_well_formed: 98.2
+  error_outputs: 8
+  num_malformed_responses: 8
+  num_with_malformed_responses: 4
+  user_asks: 32
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 0
+  test_timeouts: 9
+  total_tests: 225
+  command: aider --model gemini/gemini-exp-1206
+  date: 2024-12-22
+  versions: 0.69.2.dev
+  seconds_per_case: 45.5
+  total_cost: 0.0000
+  
+- dirname: 2024-12-22-20-08-13--gemini-2.0-flash-exp-polyglot-whole
+  test_cases: 225
+  model: gemini-2.0-flash-exp
+  edit_format: whole
+  commit_hash: b1bc2f8
+  pass_rate_1: 11.6
+  pass_rate_2: 22.2
+  pass_num_1: 26
+  pass_num_2: 50
+  percent_cases_well_formed: 100.0
+  error_outputs: 1
+  num_malformed_responses: 0
+  num_with_malformed_responses: 0
+  user_asks: 9
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 1
+  test_timeouts: 8
+  total_tests: 225
+  command: aider --model gemini/gemini-2.0-flash-exp
+  date: 2024-12-22
+  versions: 0.69.2.dev
+  seconds_per_case: 12.2
+  total_cost: 0.0000

aider/website/_data/polyglot_leaderboard.yml

@@ -0,0 +1,546 @@
+- dirname: 2024-12-21-18-41-18--polyglot-gpt-4o-mini
+  test_cases: 225
+  model: gpt-4o-mini-2024-07-18
+  edit_format: whole
+  commit_hash: a755079-dirty
+  pass_rate_1: 0.9
+  pass_rate_2: 3.6
+  pass_num_1: 2
+  pass_num_2: 8
+  percent_cases_well_formed: 100.0
+  error_outputs: 0
+  num_malformed_responses: 0
+  num_with_malformed_responses: 0
+  user_asks: 36
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 0
+  test_timeouts: 3
+  total_tests: 225
+  command: aider --model gpt-4o-mini-2024-07-18
+  date: 2024-12-21
+  versions: 0.69.2.dev
+  seconds_per_case: 17.3
+  total_cost: 0.3236
+
+- dirname: 2025-01-17-19-44-33--sonnet-baseline-jan-17
+  test_cases: 225
+  model: claude-3-5-sonnet-20241022
+  edit_format: diff
+  commit_hash: 6451d59
+  pass_rate_1: 22.2
+  pass_rate_2: 51.6
+  pass_num_1: 50
+  pass_num_2: 116
+  percent_cases_well_formed: 99.6
+  error_outputs: 2
+  num_malformed_responses: 1
+  num_with_malformed_responses: 1
+  user_asks: 11
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 1
+  test_timeouts: 8
+  total_tests: 225
+  command: aider --model claude-3-5-sonnet-20241022
+  date: 2025-01-17
+  versions: 0.71.2.dev
+  seconds_per_case: 21.4
+  total_cost: 14.4063
+  
+- dirname: 2024-12-30-20-57-12--gpt-4o-2024-11-20-ex-as-sys
+  test_cases: 225
+  model: gpt-4o-2024-11-20
+  edit_format: diff
+  commit_hash: 09ee197-dirty
+  pass_rate_1: 4.9
+  pass_rate_2: 18.2
+  pass_num_1: 11
+  pass_num_2: 41
+  percent_cases_well_formed: 95.1
+  error_outputs: 12
+  num_malformed_responses: 12
+  num_with_malformed_responses: 11
+  user_asks: 53
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 0
+  test_timeouts: 12
+  total_tests: 225
+  command: aider --model gpt-4o-2024-11-20
+  date: 2024-12-30
+  versions: 0.70.1.dev
+  seconds_per_case: 12.1
+  total_cost: 6.7351
+  
+- dirname: 2024-12-30-20-44-54--gpt4o-ex-as-sys-clean-prompt
+  test_cases: 225
+  model: gpt-4o-2024-08-06
+  edit_format: diff
+  commit_hash: 09ee197-dirty
+  pass_rate_1: 4.9
+  pass_rate_2: 23.1
+  pass_num_1: 11
+  pass_num_2: 52
+  percent_cases_well_formed: 94.2
+  error_outputs: 21
+  num_malformed_responses: 21
+  num_with_malformed_responses: 13
+  user_asks: 65
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 0
+  test_timeouts: 3
+  total_tests: 225
+  command: aider --model gpt-4o-2024-08-06
+  date: 2024-12-30
+  versions: 0.70.1.dev
+  seconds_per_case: 16.0
+  total_cost: 7.0286
+  
+- dirname: 2024-12-21-19-23-03--polyglot-o1-hard-diff
+  test_cases: 224
+  model: o1-2024-12-17 (high)
+  edit_format: diff
+  commit_hash: a755079-dirty
+  pass_rate_1: 23.7
+  pass_rate_2: 61.7
+  pass_num_1: 53
+  pass_num_2: 139
+  percent_cases_well_formed: 91.5
+  error_outputs: 25
+  num_malformed_responses: 24
+  num_with_malformed_responses: 19
+  user_asks: 16
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 0
+  test_timeouts: 2
+  total_tests: 225
+  command: aider --model openrouter/openai/o1
+  date: 2024-12-21
+  versions: 0.69.2.dev
+  seconds_per_case: 133.2
+  total_cost: 186.4958
+
+- dirname: 2024-12-21-20-56-21--polyglot-deepseek-diff
+  test_cases: 225
+  model: DeepSeek Chat V2.5
+  edit_format: diff
+  commit_hash: a755079-dirty
+  pass_rate_1: 5.3
+  pass_rate_2: 17.8
+  pass_num_1: 12
+  pass_num_2: 40
+  percent_cases_well_formed: 92.9
+  error_outputs: 42
+  num_malformed_responses: 37
+  num_with_malformed_responses: 16
+  user_asks: 23
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 5
+  test_timeouts: 5
+  total_tests: 225
+  command: aider --model deepseek/deepseek-chat
+  date: 2024-12-21
+  versions: 0.69.2.dev
+  seconds_per_case: 184.0
+  total_cost: 0.5101
+
+- dirname: 2024-12-21-21-46-27--polyglot-haiku-diff
+  test_cases: 225
+  model: claude-3-5-haiku-20241022
+  edit_format: diff
+  commit_hash: a755079-dirty
+  pass_rate_1: 7.1
+  pass_rate_2: 28.0
+  pass_num_1: 16
+  pass_num_2: 63
+  percent_cases_well_formed: 91.1
+  error_outputs: 31
+  num_malformed_responses: 30
+  num_with_malformed_responses: 20
+  user_asks: 13
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 1
+  test_timeouts: 9
+  total_tests: 225
+  command: aider --model claude-3-5-haiku-20241022
+  date: 2024-12-21
+  versions: 0.69.2.dev
+  seconds_per_case: 31.8
+  total_cost: 6.0583
+
+- dirname: 2024-12-22-13-22-32--polyglot-qwen-diff
+  test_cases: 225
+  model: Qwen2.5-Coder-32B-Instruct
+  edit_format: diff
+  commit_hash: 6d7e8be-dirty
+  pass_rate_1: 4.4
+  pass_rate_2: 8.0
+  pass_num_1: 10
+  pass_num_2: 18
+  percent_cases_well_formed: 71.6
+  error_outputs: 158
+  num_malformed_responses: 148
+  num_with_malformed_responses: 64
+  user_asks: 132
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 1
+  test_timeouts: 2
+  total_tests: 225
+  command: "aider --model openai/Qwen/Qwen2.5-Coder-32B-Instruct # via hyperbolic"
+  date: 2024-12-22
+  versions: 0.69.2.dev
+  seconds_per_case: 84.4
+  total_cost: 0.0000
+
+- dirname: 2024-12-22-21-26-35--polyglot-o1mini-whole
+  test_cases: 225
+  model: o1-mini-2024-09-12
+  edit_format: whole
+  commit_hash: 37df899
+  pass_rate_1: 5.8
+  pass_rate_2: 32.9
+  pass_num_1: 13
+  pass_num_2: 74
+  percent_cases_well_formed: 96.9
+  error_outputs: 8
+  num_malformed_responses: 8
+  num_with_malformed_responses: 7
+  user_asks: 27
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 0
+  test_timeouts: 3
+  total_tests: 225
+  command: aider --model o1-mini
+  date: 2024-12-22
+  versions: 0.69.2.dev
+  seconds_per_case: 34.7
+  total_cost: 18.5770
+
+- dirname: 2024-12-22-18-43-25--gemini-exp-1206-polyglot-whole-2
+  test_cases: 225
+  model: gemini-exp-1206
+  edit_format: whole
+  commit_hash: b1bc2f8
+  pass_rate_1: 19.6
+  pass_rate_2: 38.2
+  pass_num_1: 44
+  pass_num_2: 86
+  percent_cases_well_formed: 98.2
+  error_outputs: 8
+  num_malformed_responses: 8
+  num_with_malformed_responses: 4
+  user_asks: 32
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 0
+  test_timeouts: 9
+  total_tests: 225
+  command: aider --model gemini/gemini-exp-1206
+  date: 2024-12-22
+  versions: 0.69.2.dev
+  seconds_per_case: 45.5
+  total_cost: 0.0000
+  
+- dirname: 2024-12-22-20-08-13--gemini-2.0-flash-exp-polyglot-whole
+  test_cases: 225
+  model: gemini-2.0-flash-exp
+  edit_format: whole
+  commit_hash: b1bc2f8
+  pass_rate_1: 11.6
+  pass_rate_2: 22.2
+  pass_num_1: 26
+  pass_num_2: 50
+  percent_cases_well_formed: 100.0
+  error_outputs: 1
+  num_malformed_responses: 0
+  num_with_malformed_responses: 0
+  user_asks: 9
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 1
+  test_timeouts: 8
+  total_tests: 225
+  command: aider --model gemini/gemini-2.0-flash-exp
+  date: 2024-12-22
+  versions: 0.69.2.dev
+  seconds_per_case: 12.2
+  total_cost: 0.0000
+  
+- dirname: 2024-12-23-01-11-56--yi-test
+  test_cases: 225
+  model: yi-lightning
+  edit_format: whole
+  commit_hash: 2b1625e
+  pass_rate_1: 5.8
+  pass_rate_2: 12.9
+  pass_num_1: 13
+  pass_num_2: 29
+  percent_cases_well_formed: 92.9
+  error_outputs: 87
+  num_malformed_responses: 72
+  num_with_malformed_responses: 16
+  user_asks: 107
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 1
+  test_timeouts: 6
+  total_tests: 225
+  command: aider --model openai/yi-lightning
+  date: 2024-12-23
+  versions: 0.69.2.dev
+  seconds_per_case: 146.7
+  total_cost: 0.0000
+
+- dirname: 2024-12-25-13-31-51--deepseekv3preview-diff2
+  test_cases: 225
+  model: DeepSeek Chat V3
+  edit_format: diff
+  commit_hash: 0a23c4a-dirty
+  pass_rate_1: 22.7
+  pass_rate_2: 48.4
+  pass_num_1: 51
+  pass_num_2: 109
+  percent_cases_well_formed: 98.7
+  error_outputs: 7
+  num_malformed_responses: 7
+  num_with_malformed_responses: 3
+  user_asks: 19
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 0
+  test_timeouts: 8
+  total_tests: 225
+  command: aider --model deepseek/deepseek-chat
+  date: 2024-12-25
+  versions: 0.69.2.dev
+  seconds_per_case: 34.8
+  total_cost: 0.3369
+
+- dirname: 2024-12-26-00-55-20--Qwen2.5-Coder-32B-Instruct
+  test_cases: 225
+  model: Qwen2.5-Coder-32B-Instruct
+  edit_format: whole
+  commit_hash: b51768b0
+  pass_rate_1: 4.9
+  pass_rate_2: 16.4
+  pass_num_1: 11
+  pass_num_2: 37
+  percent_cases_well_formed: 99.6
+  error_outputs: 1
+  num_malformed_responses: 1
+  num_with_malformed_responses: 1
+  user_asks: 33
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 0
+  test_timeouts: 6
+  total_tests: 225
+  command: aider --model openai/Qwen2.5-Coder-32B-Instruct
+  date: 2024-12-26
+  versions: 0.69.2.dev
+  seconds_per_case: 42.0
+  total_cost: 0.0000
+
+- dirname: 2025-01-13-18-17-25--codestral-whole2
+  test_cases: 225
+  model: Codestral 25.01
+  edit_format: whole
+  commit_hash: 0cba898-dirty
+  pass_rate_1: 4.0
+  pass_rate_2: 11.1
+  pass_num_1: 9
+  pass_num_2: 25
+  percent_cases_well_formed: 100.0
+  error_outputs: 0
+  num_malformed_responses: 0
+  num_with_malformed_responses: 0
+  user_asks: 47
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 0
+  test_timeouts: 4
+  total_tests: 225
+  command: aider --model mistral/codestral-latest
+  date: 2025-01-13
+  versions: 0.71.2.dev
+  seconds_per_case: 9.3
+  total_cost: 1.9834
+
+- dirname: 2025-01-20-19-11-38--ds-turns-upd-cur-msgs-fix-with-summarizer
+  test_cases: 225
+  model: DeepSeek R1
+  edit_format: diff
+  commit_hash: 5650697-dirty
+  pass_rate_1: 26.7
+  pass_rate_2: 56.9
+  pass_num_1: 60
+  pass_num_2: 128
+  percent_cases_well_formed: 96.9
+  error_outputs: 8
+  num_malformed_responses: 7
+  num_with_malformed_responses: 7
+  user_asks: 15
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 1
+  test_timeouts: 5
+  total_tests: 225
+  command: aider --model deepseek/deepseek-reasoner
+  date: 2025-01-20
+  versions: 0.71.2.dev
+  seconds_per_case: 113.7
+  total_cost: 5.4193
+
+- dirname: 2025-01-23-19-14-48--r1-architect-sonnet
+  test_cases: 225
+  model: DeepSeek R1 + claude-3-5-sonnet-20241022
+  edit_format: architect
+  commit_hash: 05a77c7
+  editor_model: claude-3-5-sonnet-20241022
+  editor_edit_format: editor-diff
+  pass_rate_1: 27.1
+  pass_rate_2: 64.0
+  pass_num_1: 61
+  pass_num_2: 144
+  percent_cases_well_formed: 100.0
+  error_outputs: 2
+  num_malformed_responses: 0
+  num_with_malformed_responses: 0
+  user_asks: 392
+  lazy_comments: 6
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 0
+  test_timeouts: 5
+  total_tests: 225
+  command: aider --architect --model r1 --editor-model sonnet
+  date: 2025-01-23
+  versions: 0.72.3.dev
+  seconds_per_case: 251.6
+  total_cost: 13.2933
+
+- dirname: 2025-01-28-16-00-03--qwen-max-2025-01-25-polyglot-diff
+  test_cases: 225
+  model: qwen-max-2025-01-25
+  edit_format: diff
+  commit_hash: ae7d459
+  pass_rate_1: 9.3
+  pass_rate_2: 21.8
+  pass_num_1: 21
+  pass_num_2: 49
+  percent_cases_well_formed: 90.2
+  error_outputs: 46
+  num_malformed_responses: 44
+  num_with_malformed_responses: 22
+  user_asks: 23
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 0
+  test_timeouts: 9
+  total_tests: 225
+  command: OPENAI_API_BASE=https://dashscope-intl.aliyuncs.com/compatible-mode/v1 aider --model openai/qwen-max-2025-01-25
+  date: 2025-01-28
+  versions: 0.72.4.dev
+  seconds_per_case: 39.5
+
+- dirname: 2025-01-31-20-27-46--o3-mini-diff2
+  test_cases: 225
+  model: o3-mini (medium)
+  edit_format: diff
+  commit_hash: 2fb517b-dirty
+  pass_rate_1: 19.1
+  pass_rate_2: 53.8
+  pass_num_1: 43
+  pass_num_2: 121
+  percent_cases_well_formed: 95.1
+  error_outputs: 28
+  num_malformed_responses: 28
+  num_with_malformed_responses: 11
+  user_asks: 17
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 0
+  test_timeouts: 2
+  total_tests: 225
+  command: aider --model o3-mini
+  date: 2025-01-31
+  versions: 0.72.4.dev
+  seconds_per_case: 47.2
+  total_cost: 8.8599
+
+- dirname: 2025-01-31-20-42-47--o3-mini-diff-high
+  test_cases: 224
+  model: o3-mini (high)
+  edit_format: diff
+  commit_hash: b0d58d1-dirty
+  pass_rate_1: 21.0
+  pass_rate_2: 60.4
+  pass_num_1: 47
+  pass_num_2: 136
+  percent_cases_well_formed: 93.3
+  error_outputs: 26
+  num_malformed_responses: 24
+  num_with_malformed_responses: 15
+  user_asks: 19
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 1
+  test_timeouts: 7
+  total_tests: 225
+  command: aider --model o3-mini --reasoning-effort high
+  date: 2025-01-31
+  versions: 0.72.4.dev
+  seconds_per_case: 124.6
+  total_cost: 18.1584
+
+- dirname: 2025-01-21-22-51-49--gemini-2.0-flash-thinking-exp-01-21-polyglot-diff
+  test_cases: 225
+  model: gemini-2.0-flash-thinking-exp-01-21
+  edit_format: diff
+  commit_hash: 843720a
+  pass_rate_1: 5.8
+  pass_rate_2: 18.2
+  pass_num_1: 13
+  pass_num_2: 41
+  percent_cases_well_formed: 77.8
+  error_outputs: 182
+  num_malformed_responses: 180
+  num_with_malformed_responses: 50
+  user_asks: 26
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 2
+  test_timeouts: 7
+  total_tests: 225
+  command: aider --model gemini/gemini-2.0-flash-thinking-exp-01-21
+  date: 2025-01-21
+  versions: 0.72.2.dev
+  seconds_per_case: 24.2
+  total_cost: 0.0000

aider/website/_data/r1_architect.yml

@@ -0,0 +1,138 @@
+
+
+
+- dirname: 2025-01-23-19-14-48--r1-architect-sonnet
+  test_cases: 225
+  model: R1+Sonnet
+  edit_format: architect
+  commit_hash: 05a77c7
+  editor_model: claude-3-5-sonnet-20241022
+  editor_edit_format: editor-diff
+  pass_rate_1: 27.1
+  pass_rate_2: 64.0
+  pass_num_1: 61
+  pass_num_2: 144
+  percent_cases_well_formed: 100.0
+  error_outputs: 2
+  num_malformed_responses: 0
+  num_with_malformed_responses: 0
+  user_asks: 392
+  lazy_comments: 6
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 0
+  test_timeouts: 5
+  total_tests: 225
+  command: aider --architect --model r1 --editor-model sonnet
+  date: 2025-01-23
+  versions: 0.72.3.dev
+  seconds_per_case: 251.6
+  total_cost: 13.2933
+
+- dirname: 2025-01-20-19-11-38--ds-turns-upd-cur-msgs-fix-with-summarizer
+  test_cases: 225
+  model: R1
+  edit_format: diff
+  commit_hash: 5650697-dirty
+  pass_rate_1: 26.7
+  pass_rate_2: 56.9
+  pass_num_1: 60
+  pass_num_2: 128
+  percent_cases_well_formed: 96.9
+  error_outputs: 8
+  num_malformed_responses: 7
+  num_with_malformed_responses: 7
+  user_asks: 15
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 1
+  test_timeouts: 5
+  total_tests: 225
+  command: aider --model r1
+  date: 2025-01-20
+  versions: 0.71.2.dev
+  seconds_per_case: 113.7
+  total_cost: 5.4193
+
+
+- dirname: 2024-12-21-19-23-03--polyglot-o1-hard-diff
+  test_cases: 224
+  model: o1
+  edit_format: diff
+  commit_hash: a755079-dirty
+  pass_rate_1: 23.7
+  pass_rate_2: 61.7
+  pass_num_1: 53
+  pass_num_2: 139
+  percent_cases_well_formed: 91.5
+  error_outputs: 25
+  num_malformed_responses: 24
+  num_with_malformed_responses: 19
+  user_asks: 16
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 0
+  test_timeouts: 2
+  total_tests: 225
+  command: aider --model o1
+  date: 2024-12-21
+  versions: 0.69.2.dev
+  seconds_per_case: 133.2
+  total_cost: 186.4958
+
+
+- dirname: 2024-12-25-13-31-51--deepseekv3preview-diff2
+  test_cases: 225
+  model: DeepSeek V3
+  edit_format: diff
+  commit_hash: 0a23c4a-dirty
+  pass_rate_1: 22.7
+  pass_rate_2: 48.4
+  pass_num_1: 51
+  pass_num_2: 109
+  percent_cases_well_formed: 98.7
+  error_outputs: 7
+  num_malformed_responses: 7
+  num_with_malformed_responses: 3
+  user_asks: 19
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 0
+  test_timeouts: 8
+  total_tests: 225
+  command: aider --model deepseek
+  date: 2024-12-25
+  versions: 0.69.2.dev
+  seconds_per_case: 34.8
+  total_cost: 0.3369
+
+
+
+- dirname: 2025-01-17-19-44-33--sonnet-baseline-jan-17
+  test_cases: 225
+  model: Sonnet
+  edit_format: diff
+  commit_hash: 6451d59
+  pass_rate_1: 22.2
+  pass_rate_2: 51.6
+  pass_num_1: 50
+  pass_num_2: 116
+  percent_cases_well_formed: 99.6
+  error_outputs: 2
+  num_malformed_responses: 1
+  num_with_malformed_responses: 1
+  user_asks: 11
+  lazy_comments: 0
+  syntax_errors: 0
+  indentation_errors: 0
+  exhausted_context_windows: 1
+  test_timeouts: 8
+  total_tests: 225
+  command: aider --model sonnet
+  date: 2025-01-17
+  versions: 0.71.2.dev
+  seconds_per_case: 21.4
+  total_cost: 14.4063

aider/website/_includes/blame.md

@@ -1,5 +1,18 @@
-<canvas id="blameChart" width="800" height="360" style="margin-top: 20px"></canvas>
-<canvas id="linesChart" width="800" height="360" style="margin-top: 20px"></canvas>
+<div class="chart-container">
+    <canvas id="blameChart" style="margin-top: 20px"></canvas>
+</div>
+<div class="chart-container">
+    <canvas id="linesChart" style="margin-top: 20px"></canvas>
+</div>
+
+<style>
+.chart-container {
+    position: relative;
+    width: 100%;
+    height: 300px;
+}
+</style>
+
 <script src="https://cdn.jsdelivr.net/npm/chart.js"></script>
 <script src="https://cdn.jsdelivr.net/npm/moment"></script>
 <script src="https://cdn.jsdelivr.net/npm/chartjs-adapter-moment"></script>
@@ -24,10 +37,17 @@ document.addEventListener('DOMContentLoaded', function () {
     var linesData = {
         labels: labels,
         datasets: [{
-            label: 'Aider\'s lines of new code',
+            label: 'Aider',
             data: [{% for row in site.data.blame %}{ x: '{{ row.end_tag }}', y: {{ row.aider_total }} },{% endfor %}],
-            backgroundColor: 'rgba(255, 99, 132, 0.8)',
-            borderColor: 'rgba(255, 99, 132, 1)',
+            backgroundColor: 'rgba(54, 162, 235, 0.8)',
+            borderColor: 'rgba(54, 162, 235, 1)',
+            borderWidth: 1
+        },
+        {
+            label: 'Human',
+            data: [{% for row in site.data.blame %}{ x: '{{ row.end_tag }}', y: {{ row.total_lines | minus: row.aider_total }} },{% endfor %}],
+            backgroundColor: 'rgba(200, 200, 200, 0.8)',
+            borderColor: 'rgba(200, 200, 200, 1)',
             borderWidth: 1
         }]
     };
@@ -36,6 +56,7 @@ document.addEventListener('DOMContentLoaded', function () {
         type: 'bar',
         data: blameData,
         options: {
+            maintainAspectRatio: false,
             scales: {
                 x: {
                     type: 'category',
@@ -85,9 +106,11 @@ document.addEventListener('DOMContentLoaded', function () {
         type: 'bar',
         data: linesData,
         options: {
+            maintainAspectRatio: false,
             scales: {
                 x: {
                     type: 'category',
+                    stacked: true,
                     title: {
                         display: true,
                         text: 'Version'
@@ -98,6 +121,7 @@ document.addEventListener('DOMContentLoaded', function () {
                     }
                 },
                 y: {
+                    stacked: true,
                     title: {
                         display: true,
                         text: 'Lines of new code'
@@ -107,12 +131,14 @@ document.addEventListener('DOMContentLoaded', function () {
             },
             plugins: {
                 legend: {
-                    display: false
+                    display: true,
+                    position: 'chartArea',
+                    reverse: true
                 },
                 tooltip: {
                     callbacks: {
                         label: function(context) {
-                            var label = 'New lines of code by aider';
+                            var label = context.dataset.label;
                             var value = context.parsed.y || 0;
                             return `${label}: ${value}`;
                         }
@@ -120,7 +146,7 @@ document.addEventListener('DOMContentLoaded', function () {
                 },
                 title: {
                     display: true,
-                    text: 'Lines of new code written by aider, by release',
+                    text: 'Lines of new code, by release',
                     font: {
                         size: 16
                     }

aider/website/_includes/edit-leaderboard.js

@@ -1,185 +0,0 @@
-document.addEventListener('DOMContentLoaded', function () {
-  var ctx = document.getElementById('editChart').getContext('2d');
-  const blueDiagonalPattern = pattern.draw('diagonal', 'rgba(54, 162, 235, 0.2)');
-  const redDiagonalPattern = pattern.draw('diagonal', 'rgba(255, 99, 132, 0.2)');
-  let displayedData = [];
-
-  const HIGHTLIGHT_MODEL = 'no no no';
-  var leaderboardData = {
-    labels: [],
-    datasets: [{
-      label: 'Percent completed correctly',
-      data: [],
-      backgroundColor: function(context) {
-        const row = allData[context.dataIndex];
-        if (row && row.edit_format === 'whole') {
-          return diagonalPattern;
-        }
-        const label = leaderboardData.labels[context.dataIndex] || '';
-        return (label && label.includes(HIGHTLIGHT_MODEL)) ? 'rgba(255, 99, 132, 0.2)' : 'rgba(54, 162, 235, 0.2)';
-      },
-      borderColor: function(context) {
-        const label = context.chart.data.labels[context.dataIndex] || '';
-        return (label && label.includes(HIGHTLIGHT_MODEL)) ? 'rgba(255, 99, 132, 1)' : 'rgba(54, 162, 235, 1)';
-      },
-      borderWidth: 1
-    }]
-  };
-
-  var allData = [];
-  {% for row in edit_sorted %}
-    allData.push({
-      model: '{{ row.model }}',
-      pass_rate_2: {{ row.pass_rate_2 }},
-      percent_cases_well_formed: {{ row.percent_cases_well_formed }},
-      edit_format: '{{ row.edit_format }}'
-    });
-  {% endfor %}
-
-  function updateChart() {
-    var selectedRows = document.querySelectorAll('tr.selected');
-    var showAll = selectedRows.length === 0;
-
-    displayedData = [];
-    leaderboardData.labels = [];
-    leaderboardData.datasets[0].data = [];
-
-    allData.forEach(function(row, index) {
-      var rowElement = document.getElementById('edit-row-' + index);
-      if (showAll) {
-        rowElement.classList.remove('selected');
-      }
-      if (showAll || rowElement.classList.contains('selected')) {
-        displayedData.push(row);
-        leaderboardData.labels.push(row.model);
-        leaderboardData.datasets[0].data.push(row.pass_rate_2);
-      }
-    });
-
-    leaderboardChart.update();
-    leaderboardChart.render();
-  }
-
-  // Use displayedData in the backgroundColor callback instead of allData
-  leaderboardData.datasets[0].backgroundColor = function(context) {
-    const row = displayedData[context.dataIndex];
-    const label = leaderboardData.labels[context.dataIndex] || '';
-    if (label && label.includes(HIGHTLIGHT_MODEL)) {
-      if (row && row.edit_format === 'whole') return redDiagonalPattern;
-      else return 'rgba(255, 99, 132, 0.2)';
-    } else if (row && row.edit_format === 'whole') {
-      return blueDiagonalPattern;
-    } else {
-      return 'rgba(54, 162, 235, 0.2)';
-    }
-  };
-
-  var tableBody = document.querySelector('table tbody');
-  allData.forEach(function(row, index) {
-    var tr = tableBody.children[index];
-    tr.id = 'edit-row-' + index;
-    tr.style.cursor = 'pointer';
-    tr.onclick = function() {
-      this.classList.toggle('selected');
-      updateChart();
-    };
-  });
-
-  var leaderboardChart = new Chart(ctx, {
-    type: 'bar',
-    data: leaderboardData,
-    options: {
-      plugins: {
-        legend: {
-          display: true,
-          labels: {
-            generateLabels: function(chart) {
-              return [
-                {
-                  text: 'Diff-like format',
-                  fillStyle: 'rgba(54, 162, 235, 0.2)',
-                  strokeStyle: 'rgba(54, 162, 235, 1)',
-                  lineWidth: 1
-                },
-                {
-                  text: 'Whole format',
-                  fillStyle: blueDiagonalPattern,
-                  strokeStyle: 'rgba(54, 162, 235, 1)',
-                  lineWidth: 1
-                }
-              ];
-            }
-          }
-        }
-      },
-      scales: {
-        y: {
-          beginAtZero: true,
-          title: {
-            display: true,
-            text: 'Percent completed correctly'
-          }
-        },
-        x: {
-          ticks: {
-            callback: function(value, index) {
-              const label = this.getLabelForValue(value);
-              if (label.length <= "claude-3-5-sonnet".length) {
-                return label;
-              }
-              
-              // Find all possible split positions
-              const splitPositions = [];
-              for (let i = 0; i < label.length; i++) {
-                if (label[i] === '-' || label[i] === ' ') {
-                  splitPositions.push(i);
-                }
-              }
-              
-              if (splitPositions.length === 0) {
-                return label;
-              }
-              
-              // Find split position closest to middle
-              const middle = label.length / 2;
-              const splitIndex = splitPositions.reduce((closest, current) => {
-                return Math.abs(current - middle) < Math.abs(closest - middle) ? current : closest;
-              });
-              
-              return [
-                label.slice(0, splitIndex),
-                label.slice(splitIndex + 1)
-              ];
-            }
-          }
-        }
-      }
-    }
-  });
-
-  updateChart();
-  
-  // Add search functionality for edit table
-  document.getElementById('editSearchInput').addEventListener('keyup', function() {
-    var searchWords = this.value.toLowerCase().split(' ').filter(word => word.length > 0);
-    var tableBody = document.querySelector('table:first-of-type tbody');
-    var rows = tableBody.getElementsByTagName('tr');
-    
-    displayedData = [];
-    leaderboardData.labels = [];
-    leaderboardData.datasets[0].data = [];
-    
-    for (var i = 0; i < rows.length; i++) {
-      var rowText = rows[i].textContent;
-      if (searchWords.every(word => rowText.toLowerCase().includes(word))) {
-        rows[i].style.display = '';
-        displayedData.push(allData[i]);
-        leaderboardData.labels.push(allData[i].model);
-        leaderboardData.datasets[0].data.push(allData[i].pass_rate_2);
-      } else {
-        rows[i].style.display = 'none';
-      }
-    }
-    leaderboardChart.update();
-  });
-});

aider/website/_includes/env-keys-tip.md

@@ -1,5 +0,0 @@
-{: .tip }
-All API keys can be stored in a
-[.env file](/docs/config/dotenv.html#storing-llm-keys)
-or in a [YAML config file](/docs/config/aider_conf.html#storing-llm-keys).
-

aider/website/_includes/get-started.md

@@ -1,17 +1,25 @@
 
-You can get started quickly like this:
+If you already have python 3.8-3.13 installed, you can get started quickly like this:
 
-```
-python -m pip install -U aider-chat
-
-# Change directory into a git repo
-cd /to/your/git/repo
-
-# Work with Claude 3.5 Sonnet on your repo
-export ANTHROPIC_API_KEY=your-key-goes-here
-aider
-
-# Work with GPT-4o on your repo
-export OPENAI_API_KEY=your-key-goes-here
-aider 
+```bash
+python -m pip install aider-install
+aider-install
+
+# Change directory into your code base
+cd /to/your/project
+
+# Work with DeepSeek via DeepSeek's API
+aider --model deepseek --api-key deepseek=your-key-goes-here
+
+# Work with Claude 3.5 Sonnet via Anthropic's API
+aider --model sonnet --api-key anthropic=your-key-goes-here
+
+# Work with GPT-4o via OpenAI's API
+aider --model gpt-4o --api-key openai=your-key-goes-here
+
+# Work with Sonnet via OpenRouter's API
+aider --model openrouter/anthropic/claude-3.5-sonnet --api-key openrouter=your-key-goes-here
+
+# Work with DeepSeek via OpenRouter's API
+aider --model openrouter/deepseek/deepseek-chat --api-key openrouter=your-key-goes-here
 ```

aider/website/_includes/head_custom.html

@@ -19,6 +19,7 @@
 <meta name="msapplication-TileColor" content="#da532c">
 <meta name="theme-color" content="#ffffff">
 
+{% if site.analytics.enabled %}
 <!-- Cookie Consent -->
 <link rel="stylesheet" type="text/css" href="https://cdn.jsdelivr.net/npm/cookieconsent@3/build/cookieconsent.min.css" />
 <script src="https://cdn.jsdelivr.net/npm/cookieconsent@3/build/cookieconsent.min.js" data-cfasync="false"></script>
@@ -72,9 +73,10 @@ window.addEventListener('load', function(){
 // PostHog initialization function
 function initPostHog() {
     !function(t,e){var o,n,p,r;e.__SV||(window.posthog=e,e._i=[],e.init=function(i,s,a){function g(t,e){var o=e.split(".");2==o.length&&(t=t[o[0]],e=o[1]),t[e]=function(){t.push([e].concat(Array.prototype.slice.call(arguments,0)))}}(p=t.createElement("script")).type="text/javascript",p.crossOrigin="anonymous",p.async=!0,p.src=s.api_host.replace(".i.posthog.com","-assets.i.posthog.com")+"/static/array.js",(r=t.getElementsByTagName("script")[0]).parentNode.insertBefore(p,r);var u=e;for(void 0!==a?u=e[a]=[]:a="posthog",u.people=u.people||[],u.toString=function(t){var e="posthog";return"posthog"!==a&&(e+="."+a),t||(e+=" (stub)"),e},u.people.toString=function(){return u.toString(1)+".people (stub)"},o="init capture register register_once register_for_session unregister unregister_for_session getFeatureFlag getFeatureFlagPayload isFeatureEnabled reloadFeatureFlags updateEarlyAccessFeatureEnrollment getEarlyAccessFeatures on onFeatureFlags onSessionId getSurveys getActiveMatchingSurveys renderSurvey canRenderSurvey getNextSurveyStep identify setPersonProperties group resetGroups setPersonPropertiesForFlags resetPersonPropertiesForFlags setGroupPropertiesForFlags resetGroupPropertiesForFlags reset get_distinct_id getGroups get_session_id get_session_replay_url alias set_config startSessionRecording stopSessionRecording sessionRecordingStarted captureException loadToolbar get_property getSessionProperty createPersonProfile opt_in_capturing opt_out_capturing has_opted_in_capturing has_opted_out_capturing clear_opt_in_out_capturing debug".split(" "),n=0;n<o.length;n++)g(u,o[n]);e._i.push([i,s,a])},e.__SV=1)}(document,window.posthog||[]);
-    posthog.init('phc_99T7muzafUMMZX15H8XePbMSreEUzahHbtWjy3l5Qbv', {
-        api_host:'https://us.i.posthog.com',
+    posthog.init('{{ site.analytics.posthog_key }}', {
+        api_host: '{{ site.analytics.posthog_host }}',
         person_profiles: 'identified_only'
     })
 }
 </script>
+{% endif %}

aider/website/_includes/keys.md

@@ -0,0 +1,4 @@
+{: .tip :}
+See the
+[API key configuration docs](/docs/config/api-keys.html)
+for information on how to configure and store your API keys.

aider/website/_includes/leaderboard.js

@@ -0,0 +1,190 @@
+document.addEventListener('DOMContentLoaded', function () {
+  var ctx = document.getElementById('editChart').getContext('2d');
+  const blueDiagonalPattern = pattern.draw('diagonal', 'rgba(54, 162, 235, 0.2)');
+  const redDiagonalPattern = pattern.draw('diagonal', 'rgba(255, 99, 132, 0.2)');
+  let displayedData = [];
+
+  const HIGHLIGHT_MODEL = '{{ highlight_model | default: "no no no" }}';
+  var leaderboardData = {
+    labels: [],
+    datasets: [{
+      label: 'Percent completed correctly',
+      data: [],
+      backgroundColor: function(context) {
+        const row = allData[context.dataIndex];
+        if (row && row.edit_format === 'whole') {
+          return diagonalPattern;
+        }
+        const label = leaderboardData.labels[context.dataIndex] || '';
+        return (label && label.includes(HIGHLIGHT_MODEL)) ? 'rgba(255, 99, 132, 0.2)' : 'rgba(54, 162, 235, 0.2)';
+      },
+      borderColor: function(context) {
+        const label = context.chart.data.labels[context.dataIndex] || '';
+        return (label && label.includes(HIGHLIGHT_MODEL)) ? 'rgba(255, 99, 132, 1)' : 'rgba(54, 162, 235, 1)';
+      },
+      borderWidth: 1
+    }]
+  };
+
+  var allData = [];
+  {% for row in data_source %}
+    allData.push({
+      model: '{{ row.model }}',
+      pass_rate: {{ row[pass_rate_field] }},
+      percent_cases_well_formed: {{ row.percent_cases_well_formed }},
+      edit_format: '{{ row.edit_format | default: "diff" }}'
+    });
+  {% endfor %}
+
+  function updateChart() {
+    var selectedRows = document.querySelectorAll('tr.selected');
+    var showAll = selectedRows.length === 0;
+
+    displayedData = [];
+    leaderboardData.labels = [];
+    leaderboardData.datasets[0].data = [];
+
+    allData.forEach(function(row, index) {
+      var rowElement = document.getElementById('edit-row-' + index);
+      if (showAll) {
+        rowElement.classList.remove('selected');
+      }
+      if (showAll || rowElement.classList.contains('selected')) {
+        displayedData.push(row);
+        leaderboardData.labels.push(row.model);
+        leaderboardData.datasets[0].data.push(row.pass_rate);
+      }
+    });
+
+    leaderboardChart.update();
+    leaderboardChart.render();
+  }
+
+  // Use displayedData in the backgroundColor callback instead of allData
+  leaderboardData.datasets[0].backgroundColor = function(context) {
+    const row = displayedData[context.dataIndex];
+    const label = leaderboardData.labels[context.dataIndex] || '';
+    if (label && label.includes(HIGHLIGHT_MODEL)) {
+      if (row && row.edit_format === 'whole') return redDiagonalPattern;
+      else return 'rgba(255, 99, 132, 0.2)';
+    } else if (row && row.edit_format === 'whole') {
+      return blueDiagonalPattern;
+    } else {
+      return 'rgba(54, 162, 235, 0.2)';
+    }
+  };
+
+  var tableBody = document.querySelector('table tbody');
+  allData.forEach(function(row, index) {
+    var tr = tableBody.children[index];
+    if (!tr) {
+      // If the row doesn't exist, create it
+      tr = document.createElement('tr');
+      tableBody.appendChild(tr);
+    }
+    tr.id = 'edit-row-' + index;
+    tr.style.cursor = 'pointer';
+    tr.onclick = function() {
+      this.classList.toggle('selected');
+      updateChart();
+    };
+  });
+
+  var leaderboardChart = new Chart(ctx, {
+    type: 'bar',
+    data: leaderboardData,
+    options: {
+      plugins: {
+        legend: {
+          display: {% if show_legend == false %}false{% else %}true{% endif %},
+          labels: {
+            generateLabels: function(chart) {
+              return [
+                {
+                  text: 'Diff-like format',
+                  fillStyle: 'rgba(54, 162, 235, 0.2)',
+                  strokeStyle: 'rgba(54, 162, 235, 1)',
+                  lineWidth: 1
+                },
+                {
+                  text: 'Whole format',
+                  fillStyle: blueDiagonalPattern,
+                  strokeStyle: 'rgba(54, 162, 235, 1)',
+                  lineWidth: 1
+                }
+              ];
+            }
+          }
+        }
+      },
+      scales: {
+        y: {
+          beginAtZero: true,
+          title: {
+            display: true,
+            text: 'Percent completed correctly'
+          }
+        },
+        x: {
+          ticks: {
+            callback: function(value, index) {
+              const label = this.getLabelForValue(value);
+              if (label.length <= "claude-3-5-sonnet".length) {
+                return label;
+              }
+              
+              // Find all possible split positions
+              const splitPositions = [];
+              for (let i = 0; i < label.length; i++) {
+                if (label[i] === '-' || label[i] === ' ') {
+                  splitPositions.push(i);
+                }
+              }
+              
+              if (splitPositions.length === 0) {
+                return label;
+              }
+              
+              // Find split position closest to middle
+              const middle = label.length / 2;
+              const splitIndex = splitPositions.reduce((closest, current) => {
+                return Math.abs(current - middle) < Math.abs(closest - middle) ? current : closest;
+              });
+              
+              return [
+                label.slice(0, splitIndex),
+                label.slice(splitIndex + 1)
+              ];
+            }
+          }
+        }
+      }
+    }
+  });
+
+  updateChart();
+  
+  // Add search functionality for edit table
+  document.getElementById('editSearchInput').addEventListener('keyup', function() {
+    var searchWords = this.value.toLowerCase().split(' ').filter(word => word.length > 0);
+    var tableBody = document.querySelector('table:first-of-type tbody');
+    var rows = tableBody.getElementsByTagName('tr');
+    
+    displayedData = [];
+    leaderboardData.labels = [];
+    leaderboardData.datasets[0].data = [];
+    
+    for (var i = 0; i < rows.length; i++) {
+      var rowText = rows[i].textContent;
+      if (searchWords.every(word => rowText.toLowerCase().includes(word))) {
+        rows[i].style.display = '';
+        displayedData.push(allData[i]);
+        leaderboardData.labels.push(allData[i].model);
+        leaderboardData.datasets[0].data.push(allData[i].pass_rate);
+      } else {
+        rows[i].style.display = 'none';
+      }
+    }
+    leaderboardChart.update();
+  });
+});

aider/website/_includes/multi-line.md

@@ -5,7 +5,10 @@ You can send long, multi-line messages in the chat in a few ways:
   - Use Meta-ENTER to start a new line without sending the message (Esc+ENTER in some environments).
   - Use `/paste` to paste text from the clipboard into the chat.
   - Use the `/editor` command to open your editor to create the next chat message. See [editor configuration docs](/docs/config/editor.html) for more info.
-
+  - Use multiline-mode, which swaps the function of Meta-Enter and Enter, so that Enter inserts a newline, and Meta-Enter submits your command. To enable multiline mode:
+    - Use the `/multiline-mode` command to toggle it during a session.
+    - Use the `--multiline` switch.
+  
 Example with a tag:
 ```
 {python

aider/website/_includes/refactor-leaderboard.js

@@ -1,90 +0,0 @@
-document.addEventListener('DOMContentLoaded', function () {
-  var ctx = document.getElementById('refacChart').getContext('2d');
-  var leaderboardData = {
-    labels: [],
-    datasets: [{
-      label: 'Percent completed correctly',
-      data: [],
-      backgroundColor: 'rgba(54, 162, 235, 0.2)',
-      borderColor: 'rgba(54, 162, 235, 1)',
-      borderWidth: 1
-    }]
-  };
-
-  var allData = [];
-  {% for row in refac_sorted %}
-    allData.push({
-      model: '{{ row.model }}',
-      pass_rate_1: {{ row.pass_rate_1 }},
-      percent_cases_well_formed: {{ row.percent_cases_well_formed }}
-    });
-  {% endfor %}
-
-  function updateChart() {
-    var selectedRows = document.querySelectorAll('tr.selected');
-    var showAll = selectedRows.length === 0;
-
-    leaderboardData.labels = [];
-    leaderboardData.datasets[0].data = [];
-
-    allData.forEach(function(row, index) {
-      var rowElement = document.getElementById('refac-row-' + index);
-      if (showAll) {
-        rowElement.classList.remove('selected');
-      }
-      if (showAll || rowElement.classList.contains('selected')) {
-        leaderboardData.labels.push(row.model);
-        leaderboardData.datasets[0].data.push(row.pass_rate_1);
-      }
-    });
-
-    leaderboardChart.update();
-  }
-
-  var tableBody = document.querySelectorAll('table tbody')[1];
-  allData.forEach(function(row, index) {
-    var tr = tableBody.children[index];
-    tr.id = 'refac-row-' + index;
-    tr.style.cursor = 'pointer';
-    tr.onclick = function() {
-      this.classList.toggle('selected');
-      updateChart();
-    };
-  });
-
-  var leaderboardChart = new Chart(ctx, {
-    type: 'bar',
-    data: leaderboardData,
-    options: {
-      scales: {
-        y: {
-          beginAtZero: true
-        }
-      }
-    }
-  });
-
-  updateChart();
-
-  // Add search functionality for refactoring table
-  document.getElementById('refacSearchInput').addEventListener('keyup', function() {
-    var searchWords = this.value.toLowerCase().split(' ').filter(word => word.length > 0);
-    var tableBody = document.querySelectorAll('table tbody')[1];
-    var rows = tableBody.getElementsByTagName('tr');
-    
-    leaderboardData.labels = [];
-    leaderboardData.datasets[0].data = [];
-    
-    for (var i = 0; i < rows.length; i++) {
-      var rowText = rows[i].textContent;
-      if (searchWords.every(word => rowText.toLowerCase().includes(word))) {
-        rows[i].style.display = '';
-        leaderboardData.labels.push(allData[i].model);
-        leaderboardData.datasets[0].data.push(allData[i].pass_rate_1);
-      } else {
-        rows[i].style.display = 'none';
-      }
-    }
-    leaderboardChart.update();
-  });
-});

aider/website/_includes/replit-pipx.md

@@ -1,6 +1,6 @@
 To use aider with pipx on replit, you can run these commands in the replit shell:
 
-```
+```bash
 pip install pipx
 pipx run aider-chat ...normal aider args...
 ```

aider/website/_includes/special-keys.md

@@ -1,49 +0,0 @@
-
-Aider has special support for providing
-OpenAI and Anthropic API keys
-via dedicated
-[command line switches](/docs/config/options.html#api-keys-and-settings)
-`--openai-api-key` and `--anthropic-api-key`.
-
-You can also set those API keys via special entries in the
-[yaml config file](/docs/config/aider_conf.html), like this:
-
-```yaml
-openai-api-key: <key>
-anthropic-api-key: <key>
-```
-
-All other LLM providers can use one of the following methods to set their
-keys:
-
-### API keys on the command line
-{: .no_toc }
-
-Use `--api-key provider=<key>` which has the effect of setting the environment variable `PROVIDER_API_KEY=<key>`. So `--api-key gemini=xxx` would set `GEMINI_API_KEY=xxx`.
-
-### API keys in a .env file
-{: .no_toc }
-
-The [.env file](/docs/config/dotenv.html)
-is a great place to set API keys and other provider API environment variables:
-
-```bash
-GEMINI_API_KEY=foo
-OPENROUTER_API_KEY=bar
-DEEPSEEK_API_KEY=baz
-```
-
-### API keys in .aider.conf.yml
-{: .no_toc }
-
-Or you can set API keys in the 
-[`.aider.conf.yml` file](/docs/config/aider_conf.html)
-via the `api-key` entry:
-
-```
-api-key:
-- gemini=foo      # Sets env var GEMINI_API_KEY=foo
-- openrouter=bar  # Sets env var OPENROUTER_API_KEY=bar
-- deepseek=baz    # Sets env var DEEPSEEK_API_KEY=baz
-```
-

aider/website/_includes/venv-pipx.md

@@ -1,7 +0,0 @@
-{: .tip }
-Using a Python 
-[virtual environment](https://docs.python.org/3/library/venv.html){:target="_blank"}
-is recommended.
-Or, you could
-[use pipx to install aider](/docs/install/pipx.html)
-once for your whole system.

aider/website/_includes/works-best.md

@@ -1,2 +1 @@
-Aider works best with GPT-4o & Claude 3.5 Sonnet and can 
-[connect to almost any LLM](https://aider.chat/docs/llms.html).
+Aider works best with Claude 3.5 Sonnet, DeepSeek V3, o1 & GPT-4o and can [connect to almost any LLM](https://aider.chat/docs/llms.html).

aider/website/_posts/2024-05-24-self-assembly.md

@@ -1,6 +1,6 @@
 ---
-title: Aider has written 7% of its own code
-excerpt: Aider has written 7% of its own code, via 600+ commits that inserted 4.8K and deleted 1.5K lines of code.
+title: Aider has written 7% of its own code (outdated, now 70%)
+excerpt: This article is quite out dated. Aider is currently writing about 70% of the new code in each release.
 highlight_image: /assets/self-assembly.jpg
 nav_exclude: true
 ---
@@ -8,13 +8,16 @@ nav_exclude: true
 <p class="post-date">{{ page.date | date: "%B %d, %Y" }}</p>
 {% endif %}
 
-# Aider has written 7% of its own code
+# Aider has written 7% of its own code (outdated, now 70%)
 
 [![self assembly](/assets/self-assembly.jpg)](https://aider.chat/assets/self-assembly.jpg)
 
 {: .note }
-This article is quite out dated. For current statistics, see
-[aider's release history](/HISTORY.html).
+This article is quite old and outdated. 
+Aider is currently writing about 70% of the new code
+in each release.
+See
+[aider's release history](/HISTORY.html) for the latest statistics.
 
 The
 [aider git repo](https://github.com/Aider-AI/aider)

aider/website/_posts/2024-12-21-polyglot.md

@@ -0,0 +1,216 @@
+---
+title: o1 tops aider's new polyglot leaderboard
+excerpt: o1 scores the top result on aider's new multi-language, more challenging coding benchmark.
+highlight_image: /assets/o1-polyglot.jpg
+draft: false
+nav_exclude: true
+---
+{% if page.date %}
+<p class="post-date">{{ page.date | date: "%B %d, %Y" }}</p>
+{% endif %}
+
+# o1 tops aider's new polyglot leaderboard
+{: .no_toc }
+
+<canvas id="editChart" width="800" height="450" style="margin-top: 20px"></canvas>
+
+OpenAI's new o1 model with "high" reasoning effort
+gets the top score on the
+new 
+[aider polyglot leaderboard](/docs/leaderboards/), significantly ahead of
+other top LLMs.
+The new polyglot benchmark uses many popular coding languages
+and was designed to be 
+*much more challenging* than aider's original
+[code editing benchmark](/docs/leaderboards/edit.html).
+This more clearly distinguishes 
+the performance of
+today's strongest coding models and
+leaves headroom for future LLMs.
+
+{: .note :}
+See the main 
+[aider leaderboard](https://aider.chat/docs/leaderboards/)
+for benchmark results from more models.
+This article only contains a snapshot
+of results at the time of publication.
+
+## The polyglot benchmark
+
+Like aider's original code editing benchmark,
+the new polyglot benchmark is based on Exercism
+coding exercises.
+
+The new polyglot benchmark:
+
+- Contains coding problems in C++, Go, Java, JavaScript, Python and Rust. 
+The old benchmark was solely based on Python exercises.
+- Focuses on the *most difficult* 225 exercises out of the 697 that
+Exercism provides for those languages.
+The old benchmark simply included all 133 Python exercises,
+regardless of difficulty.
+
+## Motivation and goals
+
+Aider's original code editing benchmark was 
+saturating as the top scores approached and then surpassed 80%.
+Sonnet's score of 84.2% was based on solving 112 of the 133
+exercises, leaving only 21 unsolved exercises.
+New champions were advancing the top score by
+solving just 1-2 more problems than the previous record.
+This made it hard to clearly 
+measure the
+difference in code editing skill between these top models.
+
+Part of the problem is that many of the original
+133 Python problems are very easy 
+and provide
+little challenge to today's frontier LLMs.
+Models as old as GPT 3.5 Turbo were able to solve half of the
+133 problems.
+Such easy problems simply inflate the benchmark scores 
+of modern LLMs without
+providing any data about which models are better or worse.
+
+The main goal for a new benchmark 
+was to re-calibrate the scale so that
+today's top coding LLMs 
+would occupy a wide range of scores between about 5% and 50%.
+This should leave headroom for future LLMs and
+make it possible to
+more clearly compare the relative performance of top models.
+
+## Designing the polyglot benchmark
+
+The new benchmark:
+
+- Tests LLMs with more coding languages, to increase diversity and source a larger pool of problems.
+- Includes just the most challenging coding problems and excludes easy problems that are solvable by most of today's top coding LLMs.
+- Includes more total coding problems, to enable more granularity of comparison.
+
+The new benchmark is based on Exercism coding problems
+from 6 of the most popular programming languages:
+
+- C++ 
+- Go 
+- Java
+- JavaScript
+- Python
+- Rust
+
+Exercism provides a total of 697 coding problems in those 6 languages.
+A set of 7 of today's top coding models each attempted all 697 of
+the Exercism problems:
+
+- Sonnet
+- Haiku
+- o1 Mini
+- DeepSeek
+- GPT-4o
+- Qwen 32B Coder Instruct
+- GPT-4o Mini
+
+Depending on the difficulty of the problems,
+a different number of solutions were found by the collection of
+7 models:
+
+| Solutions<br>found | Number of<br>problems | Cumulative number<br>of problems |
+|--------|-----------|------------|
+| 0      | 66        | 66         |
+| 1      | 61        | 127        |
+| 2      | 50        | 177        |
+| 3      | 48        | 225        |
+| 4      | 53        | 278        |
+| 5      | 71        | 349        |
+| 6      | 90        | 439        |
+| 7      | 258       | 697        |
+
+In the table above, you can see that 258 of the problems were solved
+by all 7 LLMs.
+These problems are far too easy, and wouldn't be good choices for the new benchmark.
+Instead, we need hard problems like the
+66 that none of the 7 models were able to solve.
+
+The new benchmark uses 
+the 225 problems that were solved by 3 or fewer models.
+This achieves a balance between hard and moderate problems,
+and provides a large but not excessive total pool of problems.
+It also represents a good diversity of coding languages:
+
+| Language    | Problems |
+|-------------|----------|
+| C++         | 26       |
+| Go          | 39       |
+| Java        | 47       |
+| JavaScript  | 49       |
+| Python      | 34       |
+| Rust        | 30       |
+| **Total**   | **225**  |
+
+## o1
+
+OpenAI's new o1 model established a very strong
+top score of 62% on the new benchmark.
+This still leaves 86 problems of headroom for future models
+to solve.
+Given the incredible pace of recent advancements, it
+will be interesting to see
+how long it will take for this new benchmark to saturate.
+
+## Benchmark problems
+
+The 225 coding problems are available in the
+[aider polyglot benchmark repo](https://github.com/Aider-AI/polyglot-benchmark)
+on GitHub.
+
+
+
+## Results
+
+<table style="width: 100%; max-width: 800px; margin: auto; border-collapse: collapse; box-shadow: 0 2px 4px rgba(0,0,0,0.1); font-size: 14px;">
+  <thead style="background-color: #f2f2f2;">
+    <tr>
+      <th style="padding: 8px; text-align: left;">Model</th>
+      <th style="padding: 8px; text-align: center;">Percent completed correctly</th>
+      <th style="padding: 8px; text-align: center;">Percent using correct edit format</th>
+      <th style="padding: 8px; text-align: left;">Command</th>
+      <th style="padding: 8px; text-align: center;">Edit format</th>
+    </tr>
+  </thead>
+  <tbody>
+    {% assign edit_sorted = site.data.o1_polyglot_leaderboard | sort: 'pass_rate_2' | reverse %}
+    {% for row in edit_sorted %}
+      <tr style="border-bottom: 1px solid #ddd;">
+        <td style="padding: 8px;">{{ row.model }}</td>
+        <td style="padding: 8px; text-align: center;">{{ row.pass_rate_2 }}%</td>
+        <td style="padding: 8px; text-align: center;">{{ row.percent_cases_well_formed }}%</td>
+        <td style="padding: 8px;"><code>{{ row.command }}</code></td>
+        <td style="padding: 8px; text-align: center;">{{ row.edit_format }}</td>
+      </tr>
+    {% endfor %}
+  </tbody>
+</table>
+
+<script src="https://unpkg.com/patternomaly/dist/patternomaly.js"></script>
+<script src="https://cdn.jsdelivr.net/npm/chart.js"></script>
+<script>
+{% assign data_source = edit_sorted %}
+{% assign pass_rate_field = "pass_rate_2" %}
+{% assign highlight_model = "o1-2024" %}
+{% include leaderboard.js %}
+</script>
+<style>
+  tr.selected {
+    color: #0056b3;
+  }
+  table {
+    table-layout: fixed;
+  }
+  td, th {
+    word-wrap: break-word;
+    overflow-wrap: break-word;
+  }
+  td:nth-child(3), td:nth-child(4) {
+    font-size: 12px;
+  }
+</style>

aider/website/_posts/2025-01-15-uv.md

@@ -0,0 +1,102 @@
+---
+title: Using uv as an installer
+excerpt: Reliably packaging & distributing python CLI tools is hard. Aider uses uv in novel ways to make it easy to install the aider CLI, its dependencies and python 3.12. All in an isolated env.
+draft: false
+nav_exclude: true
+---
+{% if page.date %}
+<p class="post-date">{{ page.date | date: "%B %d, %Y" }}</p>
+{% endif %}
+
+# Using uv as an installer
+{: .no_toc }
+
+It's hard to reliably
+package and distribute python command line tools
+to end users.
+Users frequently encounter challenges:
+dependency version conflicts, virtual environment management,
+needing to install python or a specific version of python, etc.
+
+Aider employs [uv](https://github.com/astral-sh/uv) 
+in a couple of novel ways to streamline the installation process:
+
+1. Install aider with
+`curl https://aider.chat/install.sh | sh` even if python isn't already installed.
+
+2. Users who have python 3.8+ installed can `pip install aider-install && aider-install`.
+
+Both methods use uv to **globally** install the `aider` command line program,
+with all of its dependencies in an **isolated environment**.
+They ensure that aider will run with **python 3.12**, and install that version
+if it is not already available.
+
+These uv install methods are especially helpful for aider, because it 
+has a large set of very specific dependencies.
+Since not all of aider's dependencies are available on all python versions,
+it requires python 3.9-3.12.
+
+Most users don't want to worry about these details --
+they just want a quick way to install and run aider.
+
+
+## One-liners
+
+Users can install aider with a shell one-liner, without even having python previously installed:
+
+```bash
+curl -LsSf https://aider.chat/install.sh | sh
+```
+
+This installs uv, then uses it to install python 3.12, 
+install the `aider` command line tool
+and update the user's shell path.
+Under the hood, it is simply a copy of 
+uv's own install script `https://astral.sh/uv/install.sh`
+with [one line added](https://github.com/Aider-AI/aider/blob/4251e976b3aa52c2a3af08da4b203d4d524c8e92/aider/website/install.sh#L1181), to install aider as a tool:
+
+```
+ensure "${_install_dir}/uv" tool install --force --python python3.12 aider-chat@latest
+```
+
+
+## aider-install
+
+The aider-install python package allows quick global installation of aider
+for users who already have python 3.8+ installed.
+It simply provides the `aider-install` command line program,
+which users just need to run once.
+
+```bash
+pip install aider-install
+aider-install
+```
+
+The `pip install aider-install` installs only two packages: 
+aider-install and the [uv python package](https://pypi.org/project/uv/).
+This ensures that uv is available
+in the user's environment.
+Everything else is installed in a stand-alone environment created by uv.
+
+When the user runs `aider-install`, it runs uv
+to install aider as a tool and update the user's shell path if needed:
+
+```bash
+uv tool install --force --python python3.12 aider-chat
+uv tool update-shell
+```
+
+
+## Benefits
+
+These uv install methods have been popular with users,
+providing a hassle free way to install aider and quickly get started.
+Installs are also extremely fast, much faster than pip or pipx installs
+even when uv is also installing python 3.12!
+
+There are also a number of benefits from the perspective of the tool developer/publisher.
+Since providing these install methods, far fewer users report dependency problems and 
+version conflicts as compared to users who `pip install aider-chat`.
+There is also less pressure to rapidly support the newest python versions, 
+since aider always installs with python 3.12.
+

aider/website/_posts/2025-01-24-r1-sonnet.md

@@ -0,0 +1,118 @@
+---
+title: R1+Sonnet set SOTA on aider's polyglot benchmark
+excerpt: R1+Sonnet has set a new SOTA on the aider polyglot benchmark. At 14X less cost compared to o1.
+highlight_image: /assets/r1-sonnet-sota.jpg
+draft: false
+nav_exclude: true
+---
+{% if page.date %}
+<p class="post-date">{{ page.date | date: "%B %d, %Y" }}</p>
+{% endif %}
+
+# R1+Sonnet set SOTA on aider's polyglot benchmark
+{: .no_toc }
+
+<canvas id="editChart" width="800" height="450" style="margin-top: 20px"></canvas>
+
+Aider supports [using a pair of models for coding](https://aider.chat/2024/09/26/architect.html):
+
+- An Architect model is asked to describe how to solve the coding problem. Thinking/reasoning models often work well in this role.
+- An Editor model is given the Architect's solution and asked to produce specific code editing instructions to apply those changes to existing source files.
+
+**R1 as architect with Sonnet as editor has set a new SOTA of 64.0%** on the 
+[aider polyglot benchmark](/2024/12/21/polyglot.html).
+They achieve this at **14X less cost** compared to the previous o1 SOTA result.
+
+o1 paired with Sonnet didn't produce better results than just using o1 alone.
+Using various other models as editor didn't seem to improve o1 or R1 versus their solo scores.
+This is in contrast to the first wave of thinking models like o1-preview and o1-mini,
+which improved when paired with many different editor models.
+
+o1 was set with reasoning effort high for these tests.
+
+## Try it
+
+Once you [install aider](https://aider.chat/docs/install.html),
+you can use aider, R1 and Sonnet like this:
+
+```bash
+export DEEPSEEK_API_KEY=<your-key>
+export ANTHROPIC_API_KEY=<your-key>
+
+aider --architect --model r1 --editor-model sonnet
+```
+
+Or if you have an [OpenRouter](https://openrouter.ai) account:
+
+```bash
+export OPENROUTER_API_KEY=<your-key>
+
+aider --architect --model openrouter/deepseek/deepseek-r1 --editor-model openrouter/anthropic/claude-3.5-sonnet
+```
+
+## Thinking output
+
+There has been 
+[some recent discussion](https://github.com/Aider-AI/aider/pull/2973)
+about extracting the `<think>` tokens from R1's responses
+and feeding them to Sonnet.
+That was an interesting experiment, for sure.
+
+To be clear, the results above are *not* using R1's thinking tokens, just the normal
+final output. 
+R1 is configured in aider's standard architect role with Sonnet as editor.
+The benchmark results that used the thinking tokens appear to be worse than
+the architect/editor results shared here.
+
+## Results
+
+<table style="width: 100%; max-width: 800px; margin: auto; border-collapse: collapse; box-shadow: 0 2px 4px rgba(0,0,0,0.1); font-size: 14px;">
+  <thead style="background-color: #f2f2f2;">
+    <tr>
+      <th style="padding: 8px; text-align: left;">Model</th>
+      <th style="padding: 8px; text-align: center;">Percent completed correctly</th>
+      <th style="padding: 8px; text-align: center;">Percent using correct edit format</th>
+      <th style="padding: 8px; text-align: left;">Command</th>
+      <th style="padding: 8px; text-align: center;">Edit format</th>
+      <th style="padding: 8px; text-align: center;">Total Cost</th>
+    </tr>
+  </thead>
+  <tbody>
+    {% assign edit_sorted = site.data.r1_architect | sort: 'pass_rate_2' | reverse %}
+    {% for row in edit_sorted %}
+      <tr style="border-bottom: 1px solid #ddd;">
+        <td style="padding: 8px;">{{ row.model }}</td>
+        <td style="padding: 8px; text-align: center;">{{ row.pass_rate_2 }}%</td>
+        <td style="padding: 8px; text-align: center;">{{ row.percent_cases_well_formed }}%</td>
+        <td style="padding: 8px;"><code>{{ row.command }}</code></td>
+        <td style="padding: 8px; text-align: center;">{{ row.edit_format }}</td>
+        <td style="padding: 8px; text-align: center;">{% if row.total_cost == 0 %}?{% else %}${{ row.total_cost | times: 1.0 | round: 2 }}{% endif %}</td>
+      </tr>
+    {% endfor %}
+  </tbody>
+</table>
+
+<script src="https://unpkg.com/patternomaly/dist/patternomaly.js"></script>
+<script src="https://cdn.jsdelivr.net/npm/chart.js"></script>
+<script>
+{% assign data_source = edit_sorted %}
+{% assign pass_rate_field = "pass_rate_2" %}
+{% assign highlight_model = "+" %}
+{% assign show_legend = false %}
+{% include leaderboard.js %}
+</script>
+<style>
+  tr.selected {
+    color: #0056b3;
+  }
+  table {
+    table-layout: fixed;
+  }
+  td, th {
+    word-wrap: break-word;
+    overflow-wrap: break-word;
+  }
+  td:nth-child(3), td:nth-child(4) {
+    font-size: 12px;
+  }
+</style>

aider/website/_posts/2025-01-28-deepseek-down.md

@@ -0,0 +1,257 @@
+---
+title: Alternative DeepSeek V3 providers
+excerpt: DeepSeek's API has been experiencing reliability issues. Here are alternative providers you can use.
+#highlight_image: /assets/deepseek-down.jpg
+draft: false
+nav_exclude: true
+---
+{% if page.date %}
+<p class="post-date">{{ page.date | date: "%B %d, %Y" }}</p>
+{% endif %}
+
+# Alternative DeepSeek V3 providers
+{: .no_toc }
+
+<canvas id="editChart" width="800" height="450" style="margin-top: 20px"></canvas>
+
+DeepSeek's API has been experiencing significant reliability issues for the past 24-48+ hours, with many users reporting downtime and overload problems.
+Their [status page](https://status.deepseek.com) notes an ongoing incident.
+
+If you're affected by these issues, several alternative providers offer access to DeepSeek V3. This article compares their performance on aider's polyglot benchmark to help you choose a reliable alternative.
+
+## Providers
+{: .no_toc }
+
+* TOC
+{:toc}
+
+## OpenRouter
+
+[OpenRouter offers many DeepSeek providers](https://openrouter.ai/deepseek/deepseek-chat/providers)
+through their unified API.
+You can use aider with OpenRouter like this:
+
+```bash
+# Set your API key using environment variables
+export OPENROUTER_API_KEY=<your-key>
+aider --model openrouter/deepseek/deepseek-chat
+
+# Or use the --api-key command line option
+aider --model openrouter/deepseek/deepseek-chat --api-key openrouter=<your-key>
+
+# Or add it to .aider.conf.yml in your home directory or project root:
+api-key:
+  - openrouter=<your-key>
+```
+
+OpenRouter automatically monitors their providers and routes requests to stable
+APIs and away from those experiencing unreliable performance.
+
+But not all providers serve the same version of open source models, and not
+all have the same privacy guarantees.
+You can control which OpenRouter providers are used to serve the model via
+[aider's model settings](https://aider.chat/docs/config/adv-model-settings.html#model-settings).
+Create a `.aider.model.settings.yml` file in your home directory or git project root with settings like this:
+
+```yaml
+- name: openrouter/deepseek/deepseek-chat
+  extra_params:
+    extra_body:
+      provider:
+        # Only use these providers, in this order
+        order: ["Novita"]
+        # Don't fall back to other providers
+        allow_fallbacks: false
+```
+
+See [OpenRouter's provider routing docs](https://openrouter.ai/docs/provider-routing) for more details.
+
+
+## Fireworks
+
+```bash
+# Set your API key using environment variables
+export FIREWORKS_API_KEY=<your-key>
+aider --model fireworks_ai/accounts/fireworks/models/deepseek-chat
+
+# Or use the --api-key command line option
+aider --model fireworks_ai/accounts/fireworks/models/deepseek-chat --api-key fireworks=<your-key>
+
+# Or add it to .aider.conf.yml in your home directory or project root:
+api-key:
+  - fireworks=<your-key>
+```
+
+Create a `.aider.model.settings.yml` file in your home directory or git project root with settings like this:
+
+```yaml
+- name: fireworks_ai/accounts/fireworks/models/deepseek-chat
+  edit_format: diff
+  weak_model_name: null
+  use_repo_map: true
+  send_undo_reply: false
+  lazy: false
+  reminder: sys
+  examples_as_sys_msg: true
+  extra_params:
+    max_tokens: 8192
+  cache_control: false
+  caches_by_default: true
+  use_system_prompt: true
+  use_temperature: true
+  streaming: true
+```
+
+
+## Hyperbolic
+
+You can use [Hyperbolic's API](https://hyperbolic.xyz) as an OpenAI-compatible provider:
+
+```bash
+# Set your API key using environment variables
+export OPENAI_API_BASE=https://api.hyperbolic.xyz/v1/
+export OPENAI_API_KEY=<your-key>
+aider --model openai/deepseek-ai/DeepSeek-V3
+
+# Or use the --api-key command line option
+aider --model openai/deepseek-ai/DeepSeek-V3 --api-key openai=<your-key>
+
+# Or add it to .aider.conf.yml in your home directory or project root:
+api-key:
+  - openai=<your-key>
+```
+
+Create a `.aider.model.settings.yml` file in your home directory or git project root with settings like this:
+
+```yaml
+- name: openai/deepseek-ai/DeepSeek-V3
+  edit_format: diff
+  weak_model_name: null
+  use_repo_map: true
+  send_undo_reply: false
+  lazy: false
+  reminder: sys
+  examples_as_sys_msg: true
+  cache_control: false
+  caches_by_default: true
+  use_system_prompt: true
+  use_temperature: true
+  streaming: true
+  editor_model_name: null
+  editor_edit_format: null
+  extra_params:
+    max_tokens: 65536
+```
+
+## Ollama
+
+You can run [DeepSeek V3 via Ollama](https://ollama.com/library/deepseek-v3).
+
+```bash
+# Pull the model
+ollama pull deepseek-v3
+
+# Start your ollama server
+ollama serve
+
+# In another terminal window...
+export OLLAMA_API_BASE=http://127.0.0.1:11434 # Mac/Linux
+setx   OLLAMA_API_BASE http://127.0.0.1:11434 # Windows, restart shell after setx
+
+aider --model ollama/deepseek-v3
+```
+
+It's important to provide model settings, especially the `num_ctx` parameter to
+set the context window.
+Ollama uses a 2k context window by default, which is very small for working with aider.
+Larger context windows will allow you to work with larger amounts of code,
+but will use memory and increase latency.
+
+Unlike most other LLM servers, Ollama does not throw an error if you submit a request that exceeds the context window. Instead, it just silently truncates the request by discarding the “oldest” messages in the chat to make it fit within the context window.
+
+So if your context window is too small, you won’t get an explicit error. The biggest symptom will be that aider says it can’t see (some of) the files you added to the chat. That’s because ollama is silently discarding them because they exceed the context window.
+
+Create a `.aider.model.settings.yml` file in your home directory or git project root with settings like this:
+
+```yaml
+- name: ollama/deepseek-v3
+  edit_format: diff
+  weak_model_name: null
+  use_repo_map: true
+  send_undo_reply: false
+  lazy: false
+  reminder: sys
+  examples_as_sys_msg: true
+  cache_control: false
+  caches_by_default: true
+  use_system_prompt: true
+  use_temperature: true
+  streaming: true
+  extra_params:
+    num_ctx: 8192 # How large a context window?
+```
+
+## Other providers
+
+You will need to properly configure aider to work with DeepSeek V3 when served
+via other providers:
+
+- Determine the `--model` name to use.
+- Provide your API key to aider.
+- Add model settings to `.aider.model.settings.yml`.
+
+
+Adapt the `.aider.model.settings.yml` shown above for Fireworks. You will need to change the `name` field to match you chosen provider's model naming scheme.
+
+See [Advanced model settings](https://aider.chat/docs/config/adv-model-settings.html#model-settings) for details about all aider model settings
+
+## Results
+
+
+<table style="width: 100%; max-width: 800px; margin: auto; border-collapse: collapse; box-shadow: 0 2px 4px rgba(0,0,0,0.1); font-size: 14px;">
+  <thead style="background-color: #f2f2f2;">
+    <tr>
+      <th style="padding: 8px; text-align: left;">Model</th>
+      <th style="padding: 8px; text-align: center;">Percent completed correctly</th>
+      <th style="padding: 8px; text-align: center;">Percent using correct edit format</th>
+      <th style="padding: 8px; text-align: left;">Command</th>
+      <th style="padding: 8px; text-align: center;">Edit format</th>
+    </tr>
+  </thead>
+  <tbody>
+    {% assign edit_sorted = site.data.deepseek-down | sort: 'pass_rate_2' | reverse %}
+    {% for row in edit_sorted %}
+      <tr style="border-bottom: 1px solid #ddd;">
+        <td style="padding: 8px;">{{ row.model }}</td>
+        <td style="padding: 8px; text-align: center;">{{ row.pass_rate_2 }}%</td>
+        <td style="padding: 8px; text-align: center;">{{ row.percent_cases_well_formed }}%</td>
+        <td style="padding: 8px;"><code>{{ row.command }}</code></td>
+        <td style="padding: 8px; text-align: center;">{{ row.edit_format }}</td>
+      </tr>
+    {% endfor %}
+  </tbody>
+</table>
+
+<script src="https://unpkg.com/patternomaly/dist/patternomaly.js"></script>
+<script src="https://cdn.jsdelivr.net/npm/chart.js"></script>
+<script>
+{% assign data_source = edit_sorted %}
+{% assign pass_rate_field = "pass_rate_2" %}
+{% assign highlight_model = "DeepSeek" %}
+{% include leaderboard.js %}
+</script>
+<style>
+  tr.selected {
+    color: #0056b3;
+  }
+  table {
+    table-layout: fixed;
+  }
+  td, th {
+    word-wrap: break-word;
+    overflow-wrap: break-word;
+  }
+  td:nth-child(3), td:nth-child(4) {
+    font-size: 12px;
+  }
+</style>

aider/website/assets/sample.aider.conf.yml

@@ -44,7 +44,7 @@
 ## Use gpt-3.5-turbo model for the main chat
 #35turbo: false
 
-## Use deepseek/deepseek-coder model for the main chat
+## Use deepseek/deepseek-chat model for the main chat
 #deepseek: false
 
 ## Use o1-mini model for the main chat
@@ -113,6 +113,9 @@
 #  - yyy
 #  - zzz
 
+## Set the reasoning_effort API parameter (default: not set)
+#reasoning-effort: xxx
+
 ## Verify the SSL cert when connecting to models (default: True)
 #verify-ssl: true
 
@@ -152,7 +155,7 @@
 ###################
 # Repomap settings:
 
-## Suggested number of tokens to use for repo map, use 0 to disable (default: 1024)
+## Suggested number of tokens to use for repo map, use 0 to disable
 #map-tokens: xxx
 
 ## Control how often the repo map is refreshed. Options: auto, always, files, manual (default: auto)
@@ -218,7 +221,7 @@
 ## Set the background color for the current item in the completion menu (default: terminal's default text color)
 #completion-menu-current-bg-color: xxx
 
-## Set the markdown code theme (default: default, other options include monokai, solarized-dark, solarized-light)
+## Set the markdown code theme (default: default, other options include monokai, solarized-dark, solarized-light, or a Pygments builtin style, see https://pygments.org/styles for available themes)
 #code-theme: default
 
 ## Show diffs when committing changes (default: False)
@@ -410,6 +413,9 @@
 ## Specify the encoding for input and output (default: utf-8)
 #encoding: utf-8
 
+## Line endings to use when writing files (default: platform)
+#line-endings: platform
+
 ## Specify the config file (default: search for .aider.conf.yml in git root, cwd or home directory)
 #config: xxx
 
@@ -422,6 +428,9 @@
 ## Enable/disable fancy input with history and completion (default: True)
 #fancy-input: true
 
+## Enable/disable multi-line input mode with Meta-Enter to submit (default: False)
+#multiline: false
+
 ## Enable/disable detection and offering to add URLs to chat (default: True)
 #detect-urls: true
 

aider/website/assets/sample.env

@@ -48,7 +48,7 @@
 ## Use gpt-3.5-turbo model for the main chat
 #AIDER_35TURBO=
 
-## Use deepseek/deepseek-coder model for the main chat
+## Use deepseek/deepseek-chat model for the main chat
 #AIDER_DEEPSEEK=
 
 ## Use o1-mini model for the main chat
@@ -102,6 +102,9 @@
 ## Add a model alias (can be used multiple times)
 #AIDER_ALIAS=
 
+## Set the reasoning_effort API parameter (default: not set)
+#AIDER_REASONING_EFFORT=
+
 ## Verify the SSL cert when connecting to models (default: True)
 #AIDER_VERIFY_SSL=true
 
@@ -141,7 +144,7 @@
 ###################
 # Repomap settings:
 
-## Suggested number of tokens to use for repo map, use 0 to disable (default: 1024)
+## Suggested number of tokens to use for repo map, use 0 to disable
 #AIDER_MAP_TOKENS=
 
 ## Control how often the repo map is refreshed. Options: auto, always, files, manual (default: auto)
@@ -207,7 +210,7 @@
 ## Set the background color for the current item in the completion menu (default: terminal's default text color)
 #AIDER_COMPLETION_MENU_CURRENT_BG_COLOR=
 
-## Set the markdown code theme (default: default, other options include monokai, solarized-dark, solarized-light)
+## Set the markdown code theme (default: default, other options include monokai, solarized-dark, solarized-light, or a Pygments builtin style, see https://pygments.org/styles for available themes)
 #AIDER_CODE_THEME=default
 
 ## Show diffs when committing changes (default: False)
@@ -381,6 +384,9 @@
 ## Specify the encoding for input and output (default: utf-8)
 #AIDER_ENCODING=utf-8
 
+## Line endings to use when writing files (default: platform)
+#AIDER_LINE_ENDINGS=platform
+
 ## Specify the .env file to load (default: .env in git root)
 #AIDER_ENV_FILE=.env
 
@@ -390,6 +396,9 @@
 ## Enable/disable fancy input with history and completion (default: True)
 #AIDER_FANCY_INPUT=true
 
+## Enable/disable multi-line input mode with Meta-Enter to submit (default: False)
+#AIDER_MULTILINE=false
+
 ## Enable/disable detection and offering to add URLs to chat (default: True)
 #AIDER_DETECT_URLS=true
 

aider/website/docs/config.md

@@ -40,5 +40,5 @@ Using an `.env` file:
 AIDER_DARK_MODE=true
 ```
 
-{% include env-keys-tip.md %}
+{% include keys.md %}
 

aider/website/docs/config/aider_conf.md

@@ -7,19 +7,17 @@ description: How to configure aider with a yaml config file.
 # YAML config file
 
 Most of aider's options can be set in an `.aider.conf.yml` file.
-Aider will look for a this file in these locations and
-load whichever is found first.
+Aider will look for a this file in these locations:
 
-- As specified with the `--config <filename>` parameter.
-- The current directory.
-- The root of your git repo.
 - Your home directory.
+- The root of your git repo.
+- The current directory.
 
-## Storing LLM keys
+If the files above exist, they will be loaded in that order. Files loaded last will take priority.
 
-{% include special-keys.md %}
+You can also specify the `--config <filename>` parameter, which will only load the one config file.
 
-{% include env-keys-tip.md %}
+{% include keys.md %}
 
 ## A note on lists
 
@@ -100,7 +98,7 @@ cog.outl("```")
 ## Use gpt-3.5-turbo model for the main chat
 #35turbo: false
 
-## Use deepseek/deepseek-coder model for the main chat
+## Use deepseek/deepseek-chat model for the main chat
 #deepseek: false
 
 ## Use o1-mini model for the main chat
@@ -169,6 +167,9 @@ cog.outl("```")
 #  - yyy
 #  - zzz
 
+## Set the reasoning_effort API parameter (default: not set)
+#reasoning-effort: xxx
+
 ## Verify the SSL cert when connecting to models (default: True)
 #verify-ssl: true
 
@@ -208,7 +209,7 @@ cog.outl("```")
 ###################
 # Repomap settings:
 
-## Suggested number of tokens to use for repo map, use 0 to disable (default: 1024)
+## Suggested number of tokens to use for repo map, use 0 to disable
 #map-tokens: xxx
 
 ## Control how often the repo map is refreshed. Options: auto, always, files, manual (default: auto)
@@ -274,7 +275,7 @@ cog.outl("```")
 ## Set the background color for the current item in the completion menu (default: terminal's default text color)
 #completion-menu-current-bg-color: xxx
 
-## Set the markdown code theme (default: default, other options include monokai, solarized-dark, solarized-light)
+## Set the markdown code theme (default: default, other options include monokai, solarized-dark, solarized-light, or a Pygments builtin style, see https://pygments.org/styles for available themes)
 #code-theme: default
 
 ## Show diffs when committing changes (default: False)
@@ -466,6 +467,9 @@ cog.outl("```")
 ## Specify the encoding for input and output (default: utf-8)
 #encoding: utf-8
 
+## Line endings to use when writing files (default: platform)
+#line-endings: platform
+
 ## Specify the config file (default: search for .aider.conf.yml in git root, cwd or home directory)
 #config: xxx
 
@@ -478,6 +482,9 @@ cog.outl("```")
 ## Enable/disable fancy input with history and completion (default: True)
 #fancy-input: true
 
+## Enable/disable multi-line input mode with Meta-Enter to submit (default: False)
+#multiline: false
+
 ## Enable/disable detection and offering to add URLs to chat (default: True)
 #detect-urls: true
 

aider/website/docs/config/api-keys.md

@@ -0,0 +1,90 @@
+---
+parent: Configuration
+nav_order: 5
+description: Setting API keys for API providers.
+---
+
+# API Keys
+
+Aider lets you specify API keys in a few ways:
+
+- On the command line
+- As environment variables
+- In a `.env` file
+- In your `.aider.conf.yml` config file
+
+---
+
+## OpenAI and Anthropic
+
+Aider has special support for providing
+OpenAI and Anthropic API keys
+via dedicated switches and configuration options.
+Settings keys for other providers works a bit differently, see below.
+
+#### Command line
+
+You can set OpenAI and Anthropic API keys via
+[command line switches](/docs/config/options.html#api-keys-and-settings)
+`--openai-api-key` and `--anthropic-api-key`.
+
+
+#### Environment variables or .env file
+
+You can also store them in environment variables or a 
+[.env file](/docs/config/dotenv.html), which also works
+for every API provider:
+
+```
+OPENAI_API_KEY=<key>
+ANTHROPIC_API_KEY=<key>
+```
+
+#### Yaml config file
+You can also set those API keys via special entries in the
+[yaml config file](/docs/config/aider_conf.html), like this:
+
+```yaml
+openai-api-key: <key>
+anthropic-api-key: <key>
+```
+
+
+---
+
+## Other API providers
+
+All other LLM providers can use one of these other methods to set their API keys.
+
+#### Command line
+{: .no_toc }
+
+Use `--api-key provider=<key>` which has the effect of setting the environment variable `PROVIDER_API_KEY=<key>`. So `--api-key gemini=xxx` would set `GEMINI_API_KEY=xxx`.
+
+#### Environment variables or .env file
+{: .no_toc }
+
+You can set API keys in environment variables.
+The [.env file](/docs/config/dotenv.html)
+is a great place to store your API keys and other provider API environment variables:
+
+```bash
+GEMINI_API_KEY=foo
+OPENROUTER_API_KEY=bar
+DEEPSEEK_API_KEY=baz
+```
+
+#### Yaml config file
+
+
+You can also set API keys in the 
+[`.aider.conf.yml` file](/docs/config/aider_conf.html)
+via the `api-key` entry:
+
+```
+api-key:
+- gemini=foo      # Sets env var GEMINI_API_KEY=foo
+- openrouter=bar  # Sets env var OPENROUTER_API_KEY=bar
+- deepseek=baz    # Sets env var DEEPSEEK_API_KEY=baz
+```
+

aider/website/docs/config/dotenv.md

@@ -20,11 +20,7 @@ Aider will look for a `.env` file in these locations:
 
 If the files above exist, they will be loaded in that order. Files loaded last will take priority.
 
-## Storing LLM keys
-
-{% include special-keys.md %}
-
-{% include env-keys-tip.md %}
+{% include keys.md %}
 
 ## Sample .env file
 
@@ -92,7 +88,7 @@ cog.outl("```")
 ## Use gpt-3.5-turbo model for the main chat
 #AIDER_35TURBO=
 
-## Use deepseek/deepseek-coder model for the main chat
+## Use deepseek/deepseek-chat model for the main chat
 #AIDER_DEEPSEEK=
 
 ## Use o1-mini model for the main chat
@@ -146,6 +142,9 @@ cog.outl("```")
 ## Add a model alias (can be used multiple times)
 #AIDER_ALIAS=
 
+## Set the reasoning_effort API parameter (default: not set)
+#AIDER_REASONING_EFFORT=
+
 ## Verify the SSL cert when connecting to models (default: True)
 #AIDER_VERIFY_SSL=true
 
@@ -185,7 +184,7 @@ cog.outl("```")
 ###################
 # Repomap settings:
 
-## Suggested number of tokens to use for repo map, use 0 to disable (default: 1024)
+## Suggested number of tokens to use for repo map, use 0 to disable
 #AIDER_MAP_TOKENS=
 
 ## Control how often the repo map is refreshed. Options: auto, always, files, manual (default: auto)
@@ -251,7 +250,7 @@ cog.outl("```")
 ## Set the background color for the current item in the completion menu (default: terminal's default text color)
 #AIDER_COMPLETION_MENU_CURRENT_BG_COLOR=
 
-## Set the markdown code theme (default: default, other options include monokai, solarized-dark, solarized-light)
+## Set the markdown code theme (default: default, other options include monokai, solarized-dark, solarized-light, or a Pygments builtin style, see https://pygments.org/styles for available themes)
 #AIDER_CODE_THEME=default
 
 ## Show diffs when committing changes (default: False)
@@ -425,6 +424,9 @@ cog.outl("```")
 ## Specify the encoding for input and output (default: utf-8)
 #AIDER_ENCODING=utf-8
 
+## Line endings to use when writing files (default: platform)
+#AIDER_LINE_ENDINGS=platform
+
 ## Specify the .env file to load (default: .env in git root)
 #AIDER_ENV_FILE=.env
 
@@ -434,6 +436,9 @@ cog.outl("```")
 ## Enable/disable fancy input with history and completion (default: True)
 #AIDER_FANCY_INPUT=true
 
+## Enable/disable multi-line input mode with Meta-Enter to submit (default: False)
+#AIDER_MULTILINE=false
+
 ## Enable/disable detection and offering to add URLs to chat (default: True)
 #AIDER_DETECT_URLS=true
 

aider/website/docs/config/model-aliases.md

@@ -13,7 +13,7 @@ Model aliases allow you to create shorthand names for models you frequently use.
 You can define aliases when launching aider using the `--alias` option:
 
 ```bash
-aider --alias "fast:gpt-3.5-turbo" --alias "smart:gpt-4"
+aider --alias "fast:gpt-4o-mini" --alias "smart:o3-mini"
 ```
 
 Multiple aliases can be defined by using the `--alias` option multiple times. Each alias definition should be in the format `alias:model-name`.
@@ -24,8 +24,8 @@ You can also define aliases in your [`.aider.conf.yml` file](https://aider.chat/
 
 ```yaml
 alias:
-  - "fast:gpt-3.5-turbo"
-  - "smart:gpt-4"
+  - "fast:gpt-4o-mini"
+  - "smart:o3-mini"
   - "hacker:claude-3-sonnet-20240229"
 ```
 
@@ -34,8 +34,8 @@ alias:
 Once defined, you can use the alias instead of the full model name:
 
 ```bash
-aider --model fast  # Uses gpt-3.5-turbo
-aider --model smart  # Uses gpt-4
+aider --model fast  # Uses gpt-4o-mini
+aider --model smart  # Uses o3-mini
 ```
 
 ## Built-in Aliases
@@ -55,9 +55,11 @@ for alias, model in sorted(MODEL_ALIASES.items()):
 - `4`: gpt-4-0613
 - `4-turbo`: gpt-4-1106-preview
 - `4o`: gpt-4o
-- `deepseek`: deepseek/deepseek-coder
+- `deepseek`: deepseek/deepseek-chat
+- `flash`: gemini/gemini-2.0-flash-exp
 - `haiku`: claude-3-5-haiku-20241022
 - `opus`: claude-3-opus-20240229
+- `r1`: deepseek/deepseek-reasoner
 - `sonnet`: claude-3-5-sonnet-20241022
 <!--[[[end]]]-->
 

aider/website/docs/config/options.md

@@ -13,9 +13,7 @@ or review them below.
 - TOC
 {:toc}
 
-## LLM keys
-
-{% include special-keys.md %}
+{% include keys.md %}
 
 ## Usage summary
 
@@ -32,7 +30,8 @@ usage: aider [-h] [--model] [--opus] [--sonnet] [--haiku] [--4]
              [--openai-api-deployment-id] [--openai-organization-id]
              [--set-env] [--api-key] [--list-models]
              [--model-settings-file] [--model-metadata-file]
-             [--alias] [--verify-ssl | --no-verify-ssl] [--timeout]
+             [--alias] [--reasoning-effort]
+             [--verify-ssl | --no-verify-ssl] [--timeout]
              [--edit-format] [--architect] [--weak-model]
              [--editor-model] [--editor-edit-format]
              [--show-model-warnings | --no-show-model-warnings]
@@ -75,9 +74,10 @@ usage: aider [-h] [--model] [--opus] [--sonnet] [--haiku] [--4]
              [--show-prompts] [--voice-format] [--voice-language]
              [--voice-input-device] [--file] [--read] [--vim]
              [--chat-language] [--yes-always] [-v] [--load]
-             [--encoding] [-c] [--env-file]
+             [--encoding] [--line-endings] [-c] [--env-file]
              [--suggest-shell-commands | --no-suggest-shell-commands]
              [--fancy-input | --no-fancy-input]
+             [--multiline | --no-multiline]
              [--detect-urls | --no-detect-urls] [--editor]
 
 ```
@@ -137,7 +137,7 @@ Aliases:
   - `-3`
 
 ### `--deepseek`
-Use deepseek/deepseek-coder model for the main chat  
+Use deepseek/deepseek-chat model for the main chat  
 Environment variable: `AIDER_DEEPSEEK`  
 
 ### `--o1-mini`
@@ -211,6 +211,10 @@ Environment variable: `AIDER_MODEL_METADATA_FILE`
 Add a model alias (can be used multiple times)  
 Environment variable: `AIDER_ALIAS`  
 
+### `--reasoning-effort VALUE`
+Set the reasoning_effort API parameter (default: not set)  
+Environment variable: `AIDER_REASONING_EFFORT`  
+
 ### `--verify-ssl`
 Verify the SSL cert when connecting to models (default: True)  
 Default: True  
@@ -276,7 +280,7 @@ Environment variable: `AIDER_CACHE_KEEPALIVE_PINGS`
 ## Repomap settings:
 
 ### `--map-tokens VALUE`
-Suggested number of tokens to use for repo map, use 0 to disable (default: 1024)  
+Suggested number of tokens to use for repo map, use 0 to disable  
 Environment variable: `AIDER_MAP_TOKENS`  
 
 ### `--map-refresh VALUE`
@@ -382,7 +386,7 @@ Set the background color for the current item in the completion menu (default: t
 Environment variable: `AIDER_COMPLETION_MENU_CURRENT_BG_COLOR`  
 
 ### `--code-theme VALUE`
-Set the markdown code theme (default: default, other options include monokai, solarized-dark, solarized-light)  
+Set the markdown code theme (default: default, other options include monokai, solarized-dark, solarized-light, or a Pygments builtin style, see https://pygments.org/styles for available themes)  
 Default: default  
 Environment variable: `AIDER_CODE_THEME`  
 
@@ -706,6 +710,11 @@ Specify the encoding for input and output (default: utf-8)
 Default: utf-8  
 Environment variable: `AIDER_ENCODING`  
 
+### `--line-endings VALUE`
+Line endings to use when writing files (default: platform)  
+Default: platform  
+Environment variable: `AIDER_LINE_ENDINGS`  
+
 ### `--config CONFIG_FILE`
 Specify the config file (default: search for .aider.conf.yml in git root, cwd or home directory)  
 Aliases:
@@ -733,6 +742,14 @@ Aliases:
   - `--fancy-input`
   - `--no-fancy-input`
 
+### `--multiline`
+Enable/disable multi-line input mode with Meta-Enter to submit (default: False)  
+Default: False  
+Environment variable: `AIDER_MULTILINE`  
+Aliases:
+  - `--multiline`
+  - `--no-multiline`
+
 ### `--detect-urls`
 Enable/disable detection and offering to add URLs to chat (default: True)  
 Default: True  

aider/website/docs/config/reasoning.md

@@ -0,0 +1,83 @@
+---
+parent: Configuration
+nav_order: 110
+description: How to configure reasoning model settings from secondary providers.
+---
+
+# Reasoning models
+
+Many 
+"reasoning" models have restrictions on how they can be used.
+They sometimes prohibit streaming, use of temperature and/or the system prompt.
+Aider is configured to work properly with these models
+when served through major provider APIs.
+
+You may need to [configure model settings](/docs/config/adv-model-settings.html)
+if you are using them through another provider
+and see errors related to temperature or system prompt.
+
+Include settings for your new provider in `.aider.model.setting.yml` file
+at the root of your project or in your home directory.
+
+## Temperature, streaming and system prompt
+
+You should find one of the existing model setting configuration entries
+for the model you are interested in, say o3-mini:
+
+```yaml
+- name: o3-mini
+  edit_format: diff
+  weak_model_name: gpt-4o-mini
+  use_repo_map: true
+  use_temperature: false             # <---
+  editor_model_name: gpt-4o
+  editor_edit_format: editor-diff
+```
+
+Pay attention to these settings, which must be set to `false`
+for certain reasoning models:
+
+- `use_temperature`
+- `streaming` 
+- `use_system_prompt`
+
+Here's an example of
+the settings to use o3-mini via Azure.
+Note that aider already has these settings pre-configured, but they
+serve as a good example of how to adapt the main model
+settings for a different provider.
+
+```yaml
+- name: azure/o3-mini
+  edit_format: diff
+  weak_model_name: azure/gpt-4o-mini
+  use_repo_map: true
+  use_temperature: false             # <---
+  editor_model_name: azure/gpt-4o
+  editor_edit_format: editor-diff
+```
+
+## Thinking tokens
+
+There is also a `remove_reasoning` setting, which takes the name of a tag.
+This is used to remove everything inside that XML tag pair.
+
+For example when using DeepSeek R1 from Fireworks, the reasoning comes back inside
+`<think>...</think>` tags, so aider's settings
+include `remove_reasoning: think` to remove that part of the response.
+
+Aider will still *display* think reasoning output, it just won't use it
+to find file editing instructions, etc.
+
+```yaml
+- name: fireworks_ai/accounts/fireworks/models/deepseek-r1
+  edit_format: diff
+  weak_model_name: fireworks_ai/accounts/fireworks/models/deepseek-v3
+  use_repo_map: true
+  extra_params:
+    max_tokens: 160000
+  use_temperature: false
+  editor_model_name: fireworks_ai/accounts/fireworks/models/deepseek-v3
+  editor_edit_format: editor-diff
+  remove_reasoning: think                 # <---
+```

aider/website/docs/faq.md

@@ -141,6 +141,18 @@ When starting a fresh aider session, you can include recent git history in the c
 
 Remember, the chat history already includes recent changes made during the current session, so this tip is most useful when starting a new aider session and you want to provide context about recent work.
 
+You can also use aider to review PR branches:
+
+```
+/run git diff one-branch..another-branch
+
+...
+
+Add 6.9k tokens of command output to the chat? (Y)es/(N)o [Yes]: Yes
+
+/ask Are there any problems with the way this change works with the FooBar class?
+```
+
 {: .tip }
 The `/git` command will not work for this purpose, as its output is not included in the chat. 
 
@@ -209,6 +221,49 @@ all the raw information being sent to/from the LLM in the conversation.
 You can also refer to the
 [instructions for installing a development version of aider](https://aider.chat/docs/install/optional.html#install-the-development-version-of-aider).
 
+## What LLMs do you use to build aider?
+
+Aider writes a lot of its own code, usually about 70% of the new code in each
+release.
+People often ask which LLMs I use with aider, when writing aider.
+Below is a table showing the models I have used recently,
+extracted from the 
+[public log](https://github.com/aider-ai/aider/blob/main/aider/website/assets/sample-analytics.jsonl)
+of my
+[aider analytics](https://aider.chat/docs/more/analytics.html).
+
+<!--[[[cog
+import sys
+sys.path.append(".")
+import scripts.my_models as my_models
+stats = my_models.collect_model_stats()
+html = my_models.format_html_table(stats)
+cog.out(html)
+]]]-->
+<style>
+table { border-collapse: collapse; width: 100%; }
+th, td { padding: 8px; text-align: left; border-bottom: 1px solid #ddd; }
+th { background-color: #f2f2f2; }
+tr:hover { background-color: #f5f5f5; }
+.right { text-align: right; }
+</style>
+<table>
+<tr><th>Model Name</th><th class='right'>Total Tokens</th><th class='right'>Percent</th></tr>
+<tr><td>claude-3-5-sonnet-20241022</td><td class='right'>654,183</td><td class='right'>45.8%</td></tr>
+<tr><td>fireworks_ai/accounts/fireworks/models/deepseek-v3</td><td class='right'>289,874</td><td class='right'>20.3%</td></tr>
+<tr><td>o3-mini</td><td class='right'>274,564</td><td class='right'>19.2%</td></tr>
+<tr><td>deepseek/deepseek-chat</td><td class='right'>97,745</td><td class='right'>6.8%</td></tr>
+<tr><td>fireworks_ai/REDACTED</td><td class='right'>45,233</td><td class='right'>3.2%</td></tr>
+<tr><td>fireworks_ai/accounts/fireworks/models/deepseek-r1</td><td class='right'>44,069</td><td class='right'>3.1%</td></tr>
+<tr><td>deepseek/deepseek-reasoner</td><td class='right'>20,223</td><td class='right'>1.4%</td></tr>
+<tr><td>gemini/REDACTED</td><td class='right'>1,859</td><td class='right'>0.1%</td></tr>
+<tr><td>ollama_chat/REDACTED</td><td class='right'>309</td><td class='right'>0.0%</td></tr>
+</table>
+
+{: .note :}
+Some models show as REDACTED, because they are new or unpopular models.
+Aider's analytics only records the names of "well known" LLMs.
+<!--[[[end]]]-->
 
 ## How are the "aider wrote xx% of code" stats computed?
 
@@ -220,6 +275,31 @@ by doing something like `git blame` on the repo,
 and counting up who wrote all the new lines of code in each release.
 Only lines in source code files are counted, not documentation or prompt files.
 
+## Why does aider sometimes stop highlighting code in its replies?
+
+Aider displays the markdown responses that are coming back from the LLM.
+Usually, the LLM will reply with code in a markdown "code block" with
+triple backtick fences, like this:
+
+````
+Here's some code:
+
+```
+print("hello")
+```
+````
+
+But if you've added files to the chat that contain triple backticks,
+aider needs to tell the LLM to use a different set of fences.
+Otherwise, the LLM can't safely include your code's triple backticks
+inside the code blocks that it returns with edits.
+Aider will use fences like `<source>...</source>` in this case.
+
+A side effect of this is that the code that aider outputs may no
+longer be properly highlighted.
+You will most often notice this if you add markdown files
+to you chats that contain code blocks.
+
 ## Why is the LLM speaking to me in an unexpected language?
 
 Aider goes to some effort to prompt the model to use the language that is configured

aider/website/docs/install.md

@@ -5,41 +5,118 @@ nav_order: 20
 description: How to install and get started pair programming with aider.
 ---
 
-# Quick start
+# Installation
+{: .no_toc }
+
+
+## Get started quickly with aider-install
 
 {% include get-started.md %}
 
-Or see the
-[full installation instructions](/docs/install/install.html)
-for more details,
-or the
-[usage instructions](https://aider.chat/docs/usage.html) to start coding with aider.
+This will install aider in its own separate python environment.
+If needed, 
+aider-install will also install a separate version of python 3.12 to use with aider.
+
+Once aider is installed,
+there are also some [optional install steps](/docs/install/optional.html).
+
+See the [usage instructions](https://aider.chat/docs/usage.html) to start coding with aider.
+
+## One-liners
+
+These one-liners will install aider, along with python 3.12 if needed.
+They are based on the 
+[uv installers](https://docs.astral.sh/uv/getting-started/installation/).
+
+#### Windows
+
+```powershell
+powershell -ExecutionPolicy ByPass -c "irm https://aider.chat/install.ps1 | iex"
+```
+
+#### Mac & Linux
+
+Use curl to download the script and execute it with sh:
+
+```bash
+curl -LsSf https://aider.chat/install.sh | sh
+```
+
+If your system doesn't have curl, you can use wget:
+
+```bash
+wget -qO- https://aider.chat/install.sh | sh
+```
+
+
+## Install with uv
+
+You can install aider with uv:
+
+```bash
+python -m pip install uv  # If you need to install uv
+uv tool install --force --python python3.12 aider-chat@latest
+```
+
+This will install uv using your existing python version 3.8-3.13,
+and use it to install aider.
+If needed, 
+uv will automatically install a separate python 3.12 to use with aider.
+
+Also see the
+[docs on other methods for installing uv itself](https://docs.astral.sh/uv/getting-started/installation/).
+
+## Install with pipx
+
+You can install aider with pipx:
+
+```bash
+python -m pip install pipx  # If you need to install pipx
+pipx install aider-chat
+```
+
+You can use pipx to install aider with python versions 3.9-3.12.
+
+Also see the
+[docs on other methods for installing pipx itself](https://pipx.pypa.io/stable/installation/).
+
+## Other install methods
+
+You can install aider with the methods described below, but one of the above
+methods is usually safer.
+
+#### Install with pip
+
+If you install with pip, you should consider
+using a 
+[virtual environment](https://docs.python.org/3/library/venv.html)
+to keep aider's dependencies separated.
+
+
+You can use pip to install aider with python versions 3.9-3.12.
+
+```bash
+# Install aider
+python -m pip install -U --upgrade-strategy only-if-needed aider-chat
+
+# To work with GPT-4o:
+aider --4o --openai-api-key sk-xxx...
+
+# To work with Claude 3.5 Sonnet:
+aider --sonnet --anthropic-api-key sk-xxx...
+```
 
 {% include python-m-aider.md %}
 
-<div class="video-container">
-  <video controls poster="/assets/install.jpg">
-    <source src="/assets/install.mp4" type="video/mp4">
-    <a href="/assets/install.mp4">Installing aider</a>
-  </video>
-</div>
+#### Installing with package managers
 
-<style>
-.video-container {
-  position: relative;
-  padding-bottom: 76.2711864407%;
-  height: 0;
-  overflow: hidden;
-}
-
-.video-container video {
-  position: absolute;
-  top: 0;
-  left: 0;
-  width: 100%;
-  height: 100%;
-}
-</style>
+It's best to install aider using one of methods
+recommended above.
+While aider is available in a number of system package managers,
+they often install aider with incorrect dependencies.
 
+## Next steps...
 
+There are some [optional install steps](/docs/install/optional.html) you could consider.
+See the [usage instructions](https://aider.chat/docs/usage.html) to start coding with aider.
 

aider/website/docs/install/codespaces.md

@@ -8,9 +8,9 @@ nav_order: 900
 
 You can use aider in GitHub Codespaces via the built-in Terminal pane.
 See below for an example,
-but you can see the 
+but you can just follow the
 [main install instructions](/docs/install.html)
-for all the details.
+inside your codespace terminal.
 
 
 <div class="video-container">

aider/website/docs/install/install.md

@@ -1,70 +0,0 @@
----
-parent: Installation
-nav_order: 10
----
-
-# Installing aider
-{: .no_toc }
-
-- TOC
-{:toc}
-
-## Python version
-
-Aider currently works with python 3.9-3.12.
-
-## Install git
-
-Make sure you have git installed.
-Here are
-[instructions for installing git in various environments](https://github.com/git-guides/install-git).
-
-## Get your API key
-
-To work with OpenAI's models like GPT-4o or GPT-3.5 you need a paid
-[OpenAI API key](https://help.openai.com/en/articles/4936850-where-do-i-find-my-secret-api-key).
-Note that this is different than being a "ChatGPT Plus" subscriber.
-
-To work with Anthropic's models like Claude 3.5 Sonnet you need a paid
-[Anthropic API key](https://docs.anthropic.com/claude/reference/getting-started-with-the-api).
-
-
-{% include venv-pipx.md %}
-
-## Mac/Linux install
-
-```
-# Install aider
-python -m pip install -U --upgrade-strategy only-if-needed aider-chat
-
-# To work with GPT-4o:
-$ aider --4o --openai-api-key sk-xxx...
-
-# To work with Claude 3.5 Sonnet:
-$ aider --sonnet --anthropic-api-key sk-xxx...
-```
-
-## Windows install
-
-```
-# Install aider
-python -m pip install -U --upgrade-strategy only-if-needed aider-chat
-
-# To work with GPT-4o:
-$ aider --4o --openai-api-key sk-xxx...
-
-# To work with Claude 3.5 Sonnet:
-$ aider --sonnet --anthropic-api-key sk-xxx...
-```
-
-{% include python-m-aider.md %}
-
-## Working with other LLMs
-
-{% include works-best.md %}
-
-## You are done!
-
-There are some [optional install steps](/docs/install/optional.html) you could consider.
-See the [usage instructions](https://aider.chat/docs/usage.html) to start coding with aider.
-

aider/website/docs/install/optional.md

@@ -11,10 +11,29 @@ The steps below are completely optional.
 - TOC
 {:toc}
 
+## Install git
 
-## Store your api keys 
+Aider works best if you have git installed.
+Here are
+[instructions for installing git in various environments](https://github.com/git-guides/install-git).
 
-You can [store your api keys in a .env file](/docs/config/dotenv.html)
+## Get your API key
+
+To work with OpenAI's models like GPT-4o or o1-preview you need a paid
+[OpenAI API key](https://help.openai.com/en/articles/4936850-where-do-i-find-my-secret-api-key).
+Note that this is different than being a "ChatGPT Plus" subscriber.
+
+To work with Anthropic's models like Claude 3.5 Sonnet you need a paid
+[Anthropic API key](https://docs.anthropic.com/claude/reference/getting-started-with-the-api).
+
+
+### Working with other LLMs
+
+{% include works-best.md %}
+
+### Store your api keys 
+
+You can [store your api keys in configuration or env files](/docs/config/api-keys.html)
 and they will be loaded automatically whenever you run aider.
 
 ## Enable Playwright 
@@ -55,13 +74,17 @@ Installing PortAudio is completely optional, but can usually be accomplished lik
 - For Linux, do `sudo apt-get install libportaudio2`
   - Some linux environments may also need `sudo apt install libasound2-plugins`
 
-## Add aider to your editor 
+## Add aider to your IDE/editor
 
-Other projects have integrated aider into some IDE/editors.
-It's not clear if they are tracking the latest
+You can use 
+[aider's `--watch-files` mode](https://aider.chat/docs/usage/watch.html)
+to integrate with any IDE or editor.
+
+There are a number of 3rd party aider plugins for various IDE/editors.
+It's not clear how well they are tracking the latest
 versions of aider,
 so it may be best to just run the latest
-aider in a terminal alongside your editor.
+aider in a terminal alongside your editor and use `--watch-files`.
 
 ### NeoVim
 
@@ -71,29 +94,22 @@ aider in a terminal alongside your editor.
 
 ### VS Code
 
-joshuavial also confirmed that aider works inside a VS Code terminal window.
-Aider detects if it is running inside VSCode and turns off pretty/color output,
-since the VSCode terminal doesn't seem to support it well.
+You can run aider inside a VS Code terminal window.
+There are a number of 3rd party 
+[aider plugins for VSCode](https://marketplace.visualstudio.com/search?term=aider%20-kodu&target=VSCode&category=All%20categories&sortBy=Relevance).
 
 ### Other editors
 
 If you are interested in creating an aider plugin for your favorite editor,
-please let me know by opening a
+please let us know by opening a
 [GitHub issue](https://github.com/Aider-AI/aider/issues).
 
 
 ## Install the development version of aider 
 
 If you want the very latest development version of aider
-you can install directly from GitHub:
+you can install it like this:
 
 ```
-python -m pip install --upgrade git+https://github.com/Aider-AI/aider.git
+aider --install-main-branch
 ```
-
-If you've git cloned the aider repository already, you can install "live" from your local copy. This is mostly useful if you are developing aider and want your current modifications to take effect immediately.
-
-```
-python -m pip install -e .
-```
-

aider/website/docs/install/pipx.md

@@ -1,33 +0,0 @@
----
-parent: Installation
-nav_order: 100
----
-
-# Install with pipx
-
-If you are using aider to work on a python project, sometimes your project will require
-specific versions of python packages which conflict with the versions that aider
-requires.
-If this happens, the `python -m pip install` command may return errors like these:
-
-```
-aider-chat 0.23.0 requires somepackage==X.Y.Z, but you have somepackage U.W.V which is incompatible.
-```
-
-You can avoid this problem by installing aider using `pipx`,
-which will install it globally on your system
-within its own python environment.
-This way you can use aider to work on any python project,
-even if that project has conflicting dependencies.
-
-Install [pipx](https://pipx.pypa.io/stable/) then just do:
-
-```
-pipx install aider-chat
-```
-
-
-## pipx on replit
-
-{% include replit-pipx.md %}
-

aider/website/docs/install/replit.md

@@ -0,0 +1,8 @@
+---
+parent: Installation
+nav_order: 900
+---
+
+### Replit
+
+{% include replit-pipx.md %}

aider/website/docs/languages.md

@@ -81,7 +81,7 @@ cog.out(get_supported_languages_md())
 | jsdoc                | .jsdoc               |          |   ✓    |
 | json                 | .json                |          |   ✓    |
 | julia                | .jl                  |          |   ✓    |
-| kotlin               | .kt                  |          |   ✓    |
+| kotlin               | .kt                  |    ✓     |   ✓    |
 | lua                  | .lua                 |          |   ✓    |
 | make                 | .mk                  |          |   ✓    |
 | objc                 | .m                   |          |   ✓    |

aider/website/docs/leaderboards/by-release-date.md

@@ -0,0 +1,10 @@
+---
+title: Scores by release date
+parent: Aider LLM Leaderboards
+nav_order: 200
+---
+
+## LLM code editing skill by model release date
+
+[![connecting to many LLMs](/assets/models-over-time.svg)](https://aider.chat/assets/models-over-time.svg)
+

aider/website/docs/leaderboards/contrib.md

@@ -0,0 +1,14 @@
+---
+parent: Aider LLM Leaderboards
+nav_order: 900
+---
+
+# Contributing results
+
+Contributions of benchmark results are welcome!
+See the
+[benchmark README](https://github.com/Aider-AI/aider/blob/main/benchmark/README.md)
+for information on running aider's code editing benchmarks.
+Submit results by opening a PR with edits to the
+[benchmark results data files](https://github.com/Aider-AI/aider/blob/main/aider/website/_data/).
+

aider/website/docs/leaderboards/edit.md

@@ -0,0 +1,133 @@
+---
+parent: Aider LLM Leaderboards
+highlight_image: /assets/leaderboard.jpg
+nav_order: 50
+description: Quantitative benchmark of basic LLM code editing skill.
+---
+
+# Code editing leaderboard
+
+
+{: .note :}
+This old
+[aider code editing leaderboard](edit.html)
+has been replaced by the
+new, much more challenging
+[polyglot leaderboard](/docs/leaderboards/).
+
+[Aider's code editing benchmark](/docs/benchmarks.html#the-benchmark) asks the LLM to edit python source files to complete 133 small coding exercises
+from Exercism. 
+This measures the LLM's coding ability, and whether it can
+write new code that integrates into existing code.
+The model also has to successfully apply all its changes to the source file without human intervention.
+
+<input type="text" id="editSearchInput" placeholder="Search..." style="width: 100%; max-width: 800px; margin: 10px auto; padding: 8px; display: block; border: 1px solid #ddd; border-radius: 4px;">
+
+<table style="width: 100%; max-width: 800px; margin: auto; border-collapse: collapse; box-shadow: 0 2px 4px rgba(0,0,0,0.1); font-size: 14px;">
+  <thead style="background-color: #f2f2f2;">
+    <tr>
+      <th style="padding: 8px; text-align: left;">Model</th>
+      <th style="padding: 8px; text-align: center;">Percent completed correctly</th>
+      <th style="padding: 8px; text-align: center;">Percent using correct edit format</th>
+      <th style="padding: 8px; text-align: left;">Command</th>
+      <th style="padding: 8px; text-align: center;">Edit format</th>
+    </tr>
+  </thead>
+  <tbody>
+    {% assign edit_sorted = site.data.edit_leaderboard | sort: 'pass_rate_2' | reverse %}
+    {% for row in edit_sorted %}
+      <tr style="border-bottom: 1px solid #ddd;">
+        <td style="padding: 8px;">{{ row.model }}</td>
+        <td style="padding: 8px; text-align: center;">{{ row.pass_rate_2 }}%</td>
+        <td style="padding: 8px; text-align: center;">{{ row.percent_cases_well_formed }}%</td>
+        <td style="padding: 8px;"><code>{{ row.command }}</code></td>
+        <td style="padding: 8px; text-align: center;">{{ row.edit_format }}</td>
+      </tr>
+    {% endfor %}
+  </tbody>
+</table>
+
+<canvas id="editChart" width="800" height="450" style="margin-top: 20px"></canvas>
+<script src="https://unpkg.com/patternomaly/dist/patternomaly.js"></script>
+<script src="https://cdn.jsdelivr.net/npm/chart.js"></script>
+<script>
+{% assign data_source = edit_sorted %}
+{% assign pass_rate_field = "pass_rate_2" %}
+{% include leaderboard.js %}
+</script>
+<style>
+  tr.selected {
+    color: #0056b3;
+  }
+  table {
+    table-layout: fixed;
+  }
+  td, th {
+    word-wrap: break-word;
+    overflow-wrap: break-word;
+  }
+  td:nth-child(3), td:nth-child(4) {
+    font-size: 12px;
+  }
+</style>
+
+
+## Notes on benchmarking results
+
+The key benchmarking results are:
+
+- **Percent completed correctly** - Measures what percentage of the coding tasks that the LLM completed successfully. To complete a task, the LLM must solve the programming assignment *and* edit the code to implement that solution.
+- **Percent using correct edit format** - Measures the percent of coding tasks where the LLM complied with the edit format specified in the system prompt. If the LLM makes edit mistakes, aider will give it feedback and ask for a fixed copy of the edit. The best models can reliably conform to the edit format, without making errors.
+
+
+## Notes on the edit format
+
+Aider uses different "edit formats" to collect code edits from different LLMs.
+The "whole" format is the easiest for an LLM to use, but it uses a lot of tokens
+and may limit how large a file can be edited.
+Models which can use one of the diff formats are much more efficient,
+using far fewer tokens.
+Models that use a diff-like format are able to 
+edit larger files with less cost and without hitting token limits.
+
+Aider is configured to use the best edit format for the popular OpenAI and Anthropic models
+and the [other models recommended on the LLM page](/docs/llms.html).
+For lesser known models aider will default to using the "whole" editing format
+since it is the easiest format for an LLM to use.
+
+## Contributing benchmark results
+
+Contributions of benchmark results are welcome!
+See the
+[benchmark README](https://github.com/Aider-AI/aider/blob/main/benchmark/README.md)
+for information on running aider's code editing benchmarks.
+Submit results by opening a PR with edits to the
+[benchmark results data files](https://github.com/Aider-AI/aider/blob/main/aider/website/_data/).
+
+
+<p class="post-date">
+By Paul Gauthier,
+last updated
+<!--[[[cog
+import subprocess
+import datetime
+
+files = [
+    'aider/website/docs/leaderboards/edit.md',
+    'aider/website/_data/edit_leaderboard.yml',
+]
+
+def get_last_modified_date(file):
+    result = subprocess.run(['git', 'log', '-1', '--format=%ct', file], capture_output=True, text=True)
+    if result.returncode == 0:
+        timestamp = int(result.stdout.strip())
+        return datetime.datetime.fromtimestamp(timestamp)
+    return datetime.datetime.min
+
+mod_dates = [get_last_modified_date(file) for file in files]
+latest_mod_date = max(mod_dates)
+cog.out(f"{latest_mod_date.strftime('%B %d, %Y.')}")
+]]]-->
+January 16, 2025.
+<!--[[[end]]]-->
+</p>

aider/website/docs/leaderboards/index.md

@@ -2,32 +2,36 @@
 highlight_image: /assets/leaderboard.jpg
 nav_order: 950
 description: Quantitative benchmarks of LLM code editing skill.
+has_children: true
 ---
 
 
 # Aider LLM Leaderboards
-{: .no_toc }
 
 Aider works best with LLMs which are good at *editing* code, not just good at writing
 code.
-To evaluate an LLM's editing skill, aider uses a pair of benchmarks that
+To evaluate an LLM's editing skill, aider uses benchmarks that
 assess a model's ability to consistently follow the system prompt
 to successfully edit code.
 
-The leaderboards below report the results from a number of popular LLMs.
+The leaderboards report the results from a number of popular LLMs.
 While [aider can connect to almost any LLM](/docs/llms.html),
 it works best with models that score well on the benchmarks.
 
-See the following sections for benchmark
-results and additional information:
-- TOC
-{:toc}
 
-## Code editing leaderboard
+## Polyglot leaderboard
 
-[Aider's code editing benchmark](/docs/benchmarks.html#the-benchmark) asks the LLM to edit python source files to complete 133 small coding exercises
+[Aider's polyglot benchmark](https://aider.chat/2024/12/21/polyglot.html#the-polyglot-benchmark) 
+asks the LLM to edit source files to complete 225 coding exercises
 from Exercism. 
-This measures the LLM's coding ability, and whether it can
+It contains exercises in many popular programming languages:
+C++, Go, Java, JavaScript, Python and Rust.
+The 225 exercises were purposely selected to be the *hardest*
+that Exercism offered in those languages, to provide
+a strong coding challenge to LLMs.
+
+This benchmark measures the LLM's coding ability in popular languages, 
+and whether it can
 write new code that integrates into existing code.
 The model also has to successfully apply all its changes to the source file without human intervention.
 
@@ -41,10 +45,11 @@ The model also has to successfully apply all its changes to the source file with
       <th style="padding: 8px; text-align: center;">Percent using correct edit format</th>
       <th style="padding: 8px; text-align: left;">Command</th>
       <th style="padding: 8px; text-align: center;">Edit format</th>
+      <th style="padding: 8px; text-align: center;">Total Cost</th>
     </tr>
   </thead>
   <tbody>
-    {% assign edit_sorted = site.data.edit_leaderboard | sort: 'pass_rate_2' | reverse %}
+    {% assign edit_sorted = site.data.polyglot_leaderboard | sort: 'pass_rate_2' | reverse %}
     {% for row in edit_sorted %}
       <tr style="border-bottom: 1px solid #ddd;">
         <td style="padding: 8px;">{{ row.model }}</td>
@@ -52,16 +57,22 @@ The model also has to successfully apply all its changes to the source file with
         <td style="padding: 8px; text-align: center;">{{ row.percent_cases_well_formed }}%</td>
         <td style="padding: 8px;"><code>{{ row.command }}</code></td>
         <td style="padding: 8px; text-align: center;">{{ row.edit_format }}</td>
+        <td style="padding: 8px; text-align: center;">{% if row.total_cost == 0 %}?{% else %}${{ row.total_cost | times: 1.0 | round: 2 }}{% endif %}</td>
       </tr>
     {% endfor %}
   </tbody>
 </table>
 
+### Aider polyglot benchmark results
+
 <canvas id="editChart" width="800" height="450" style="margin-top: 20px"></canvas>
 <script src="https://unpkg.com/patternomaly/dist/patternomaly.js"></script>
 <script src="https://cdn.jsdelivr.net/npm/chart.js"></script>
 <script>
-{% include edit-leaderboard.js %}
+{% assign data_source = edit_sorted %}
+{% assign pass_rate_field = "pass_rate_2" %}
+{% assign highlight_model = "xxxxxxxxxxx" %}
+{% include leaderboard.js %}
 </script>
 <style>
   tr.selected {
@@ -79,84 +90,7 @@ The model also has to successfully apply all its changes to the source file with
   }
 </style>
 
-## Code refactoring leaderboard
 
-[Aider's refactoring benchmark](https://github.com/Aider-AI/refactor-benchmark) asks the LLM to refactor 89 large methods from large python classes. This is a more challenging benchmark, which tests the model's ability to output long chunks of code without skipping sections or making mistakes. It was developed to provoke and measure [GPT-4 Turbo's "lazy coding" habit](/2023/12/21/unified-diffs.html).
-
-The refactoring benchmark requires a large context window to
-work with large source files.
-Therefore, results are available for fewer models.
-
-<input type="text" id="refacSearchInput" placeholder="Search..." style="width: 100%; max-width: 800px; margin: 10px auto; padding: 8px; display: block; border: 1px solid #ddd; border-radius: 4px;">
-
-<table style="width: 100%; max-width: 800px; margin: auto; border-collapse: collapse; box-shadow: 0 2px 4px rgba(0,0,0,0.1); font-size: 14px;">
-  <thead style="background-color: #f2f2f2;">
-    <tr>
-      <th style="padding: 8px; text-align: left;">Model</th>
-      <th style="padding: 8px; text-align: center;">Percent completed correctly</th>
-      <th style="padding: 8px; text-align: center;">Percent using correct edit format</th>
-      <th style="padding: 8px; text-align: left;">Command</th>
-      <th style="padding: 8px; text-align: center;">Edit format</th>
-    </tr>
-  </thead>
-  <tbody>
-    {% assign refac_sorted = site.data.refactor_leaderboard | sort: 'pass_rate_1' | reverse %}
-    {% for row in refac_sorted %}
-      <tr style="border-bottom: 1px solid #ddd;">
-        <td style="padding: 8px;">{{ row.model }}</td>
-        <td style="padding: 8px; text-align: center;">{{ row.pass_rate_1 }}%</td>
-        <td style="padding: 8px; text-align: center;">{{ row.percent_cases_well_formed }}%</td>
-        <td style="padding: 8px;"><code>{{ row.command }}</code></td>
-        <td style="padding: 8px; text-align: center;">{{ row.edit_format }}</td>
-      </tr>
-    {% endfor %}
-  </tbody>
-</table>
-
-<canvas id="refacChart" width="800" height="450" style="margin-top: 20px"></canvas>
-<script src="https://unpkg.com/patternomaly/dist/patternomaly.js"></script>
-<script src="https://cdn.jsdelivr.net/npm/chart.js"></script>
-<script>
-{% include refactor-leaderboard.js %}
-</script>
-
-
-## LLM code editing skill by model release date
-
-[![connecting to many LLMs](/assets/models-over-time.svg)](https://aider.chat/assets/models-over-time.svg)
-
-
-## Notes on benchmarking results
-
-The key benchmarking results are:
-
-- **Percent completed correctly** - Measures what percentage of the coding tasks that the LLM completed successfully. To complete a task, the LLM must solve the programming assignment *and* edit the code to implement that solution.
-- **Percent using correct edit format** - Measures the percent of coding tasks where the LLM complied with the edit format specified in the system prompt. If the LLM makes edit mistakes, aider will give it feedback and ask for a fixed copy of the edit. The best models can reliably conform to the edit format, without making errors.
-
-
-## Notes on the edit format
-
-Aider uses different "edit formats" to collect code edits from different LLMs.
-The "whole" format is the easiest for an LLM to use, but it uses a lot of tokens
-and may limit how large a file can be edited.
-Models which can use one of the diff formats are much more efficient,
-using far fewer tokens.
-Models that use a diff-like format are able to 
-edit larger files with less cost and without hitting token limits.
-
-Aider is configured to use the best edit format for the popular OpenAI and Anthropic models
-and the [other models recommended on the LLM page](/docs/llms.html).
-For lesser known models aider will default to using the "whole" editing format
-since it is the easiest format for an LLM to use.
-
-## Contributing benchmark results
-
-Contributions of benchmark results are welcome!
-See the
-[benchmark README](https://github.com/Aider-AI/aider/blob/main/benchmark/README.md)
-for information on running aider's code editing benchmarks.
-Submit results by opening a PR with edits to the
-[benchmark results data files](https://github.com/Aider-AI/aider/blob/main/aider/website/_data/).
 
 
 <p class="post-date">
@@ -168,8 +102,7 @@ import datetime
 
 files = [
     'aider/website/docs/leaderboards/index.md',
-    'aider/website/_data/edit_leaderboard.yml',
-    'aider/website/_data/refactor_leaderboard.yml'
+    'aider/website/_data/polyglot_leaderboard.yml',
 ]
 
 def get_last_modified_date(file):
@@ -183,6 +116,6 @@ mod_dates = [get_last_modified_date(file) for file in files]
 latest_mod_date = max(mod_dates)
 cog.out(f"{latest_mod_date.strftime('%B %d, %Y.')}")
 ]]]-->
-December 10, 2024.
+January 31, 2025.
 <!--[[[end]]]-->
 </p>

aider/website/docs/leaderboards/notes.md

@@ -0,0 +1,29 @@
+---
+parent: Aider LLM Leaderboards
+nav_order: 800
+---
+
+# Benchmark notes
+
+## Notes on benchmarking results
+
+The key benchmarking results are:
+
+- **Percent completed correctly** - Measures what percentage of the coding tasks that the LLM completed successfully. To complete a task, the LLM must solve the programming assignment *and* edit the code to implement that solution.
+- **Percent using correct edit format** - Measures the percent of coding tasks where the LLM complied with the edit format specified in the system prompt. If the LLM makes edit mistakes, aider will give it feedback and ask for a fixed copy of the edit. The best models can reliably conform to the edit format, without making errors.
+
+
+## Notes on the edit format
+
+Aider uses different "edit formats" to collect code edits from different LLMs.
+The "whole" format is the easiest for an LLM to use, but it uses a lot of tokens
+and may limit how large a file can be edited.
+Models which can use one of the diff formats are much more efficient,
+using far fewer tokens.
+Models that use a diff-like format are able to 
+edit larger files with less cost and without hitting token limits.
+
+Aider is configured to use the best edit format for the popular OpenAI and Anthropic models
+and the [other models recommended on the LLM page](/docs/llms.html).
+For lesser known models aider will default to using the "whole" editing format
+since it is the easiest format for an LLM to use.

aider/website/docs/leaderboards/refactor.md

@@ -0,0 +1,78 @@
+---
+parent: Aider LLM Leaderboards
+highlight_image: /assets/leaderboard.jpg
+nav_order: 100
+description: Quantitative benchmark of LLM code refactoring skill.
+---
+
+
+## Refactoring leaderboard
+
+[Aider's refactoring benchmark](https://github.com/Aider-AI/refactor-benchmark) asks the LLM to refactor 89 large methods from large python classes. This is a more challenging benchmark, which tests the model's ability to output long chunks of code without skipping sections or making mistakes. It was developed to provoke and measure [GPT-4 Turbo's "lazy coding" habit](/2023/12/21/unified-diffs.html).
+
+The refactoring benchmark requires a large context window to
+work with large source files.
+Therefore, results are available for fewer models.
+
+<input type="text" id="editSearchInput" placeholder="Search..." style="width: 100%; max-width: 800px; margin: 10px auto; padding: 8px; display: block; border: 1px solid #ddd; border-radius: 4px;">
+
+<table style="width: 100%; max-width: 800px; margin: auto; border-collapse: collapse; box-shadow: 0 2px 4px rgba(0,0,0,0.1); font-size: 14px;">
+  <thead style="background-color: #f2f2f2;">
+    <tr>
+      <th style="padding: 8px; text-align: left;">Model</th>
+      <th style="padding: 8px; text-align: center;">Percent completed correctly</th>
+      <th style="padding: 8px; text-align: center;">Percent using correct edit format</th>
+      <th style="padding: 8px; text-align: left;">Command</th>
+      <th style="padding: 8px; text-align: center;">Edit format</th>
+    </tr>
+  </thead>
+  <tbody>
+    {% assign refac_sorted = site.data.refactor_leaderboard | sort: 'pass_rate_1' | reverse %}
+    {% for row in refac_sorted %}
+      <tr style="border-bottom: 1px solid #ddd;">
+        <td style="padding: 8px;">{{ row.model }}</td>
+        <td style="padding: 8px; text-align: center;">{{ row.pass_rate_1 }}%</td>
+        <td style="padding: 8px; text-align: center;">{{ row.percent_cases_well_formed }}%</td>
+        <td style="padding: 8px;"><code>{{ row.command }}</code></td>
+        <td style="padding: 8px; text-align: center;">{{ row.edit_format }}</td>
+      </tr>
+    {% endfor %}
+  </tbody>
+</table>
+
+<canvas id="editChart" width="800" height="450" style="margin-top: 20px"></canvas>
+<script src="https://unpkg.com/patternomaly/dist/patternomaly.js"></script>
+<script src="https://cdn.jsdelivr.net/npm/chart.js"></script>
+<script>
+{% assign data_source = refac_sorted %}
+{% assign pass_rate_field = "pass_rate_1" %}
+{% include leaderboard.js %}
+</script>
+
+
+<p class="post-date">
+By Paul Gauthier,
+last updated
+<!--[[[cog
+import subprocess
+import datetime
+
+files = [
+    'aider/website/docs/leaderboards/refactor.md',
+    'aider/website/_data/refactor_leaderboard.yml',
+]
+
+def get_last_modified_date(file):
+    result = subprocess.run(['git', 'log', '-1', '--format=%ct', file], capture_output=True, text=True)
+    if result.returncode == 0:
+        timestamp = int(result.stdout.strip())
+        return datetime.datetime.fromtimestamp(timestamp)
+    return datetime.datetime.min
+
+mod_dates = [get_last_modified_date(file) for file in files]
+latest_mod_date = max(mod_dates)
+cog.out(f"{latest_mod_date.strftime('%B %d, %Y.')}")
+]]]-->
+January 16, 2025.
+<!--[[[end]]]-->
+</p>

aider/website/docs/llms.md

@@ -19,7 +19,7 @@ Aider works best with these models, which are skilled at editing code:
 - [GPT-4o](/docs/llms/openai.html)
 - [Claude 3.5 Sonnet](/docs/llms/anthropic.html)
 - [Claude 3 Opus](/docs/llms/anthropic.html)
-- [DeepSeek Coder V2](/docs/llms/deepseek.html)
+- [DeepSeek V3](/docs/llms/deepseek.html)
 
 
 ## Free models

aider/website/docs/llms/bedrock.md

@@ -53,6 +53,12 @@ To use aider installed via `pipx` with AWS Bedrock, you must add the `boto3` dep
 pipx inject aider-chat boto3
 ```
 
+You must install `boto3` dependency to aider's virtual environment installed via one-liner or uv by running
+
+```bash
+uv tool run --from aider-chat pip install boto3
+```
+
 
 ## Running Aider with Bedrock
 

aider/website/docs/llms/deepseek.md

@@ -6,7 +6,8 @@ nav_order: 500
 # DeepSeek
 
 Aider can connect to the DeepSeek.com API.
-The DeepSeek Coder V2 model has a top score on aider's code editing benchmark.
+To work with DeepSeek's models, you need to set the `DEEPSEEK_API_KEY` environment variable with your [DeepSeek API key](https://platform.deepseek.com/api_keys).  
+The DeepSeek Chat V3 model has a top score on aider's code editing benchmark.
 
 ```
 python -m pip install -U aider-chat
@@ -14,7 +15,7 @@ python -m pip install -U aider-chat
 export DEEPSEEK_API_KEY=<key> # Mac/Linux
 setx   DEEPSEEK_API_KEY <key> # Windows, restart shell after setx
 
-# Use DeepSeek Coder V2
+# Use DeepSeek Chat v3
 aider --deepseek
 ```
 

aider/website/docs/llms/ollama.md

@@ -45,7 +45,11 @@ setx   OLLAMA_API_KEY <api-key> # Windows, restart shell after setx
 [Ollama uses a 2k context window by default](https://github.com/ollama/ollama/blob/main/docs/faq.md#how-can-i-specify-the-context-window-size),
 which is very small for working with aider.
 
-Aider sets Ollama's context window to 8k by default. 
+By default, aider sets Ollama's context window 
+to be large enough for each request you send plus 8k tokens for the reply.
+
+Larger context windows may be helpful to allow larger replies from the LLM
+but will use memory and increase latency.
 If you would like
 a larger context window
 you can use a
@@ -55,14 +59,6 @@ like this:
 ```
 - name: ollama/qwen2.5-coder:32b-instruct-fp16
   extra_params:
-    num_ctx: 8192
+    num_ctx: 65536
 ```
 
-Unlike most other LLM servers, Ollama does not throw an error if you submit
-a request that exceeds the context window.
-Instead, it just silently truncates the request by discarding the "oldest" messages
-in the chat to make it fit within the context window.
-So if your context window is too small, you won't get an error.
-Aider will probably just fail to work well and experience
-a lot of 
-[file editing problems](https://aider.chat/docs/troubleshooting/edit-errors.html).

aider/website/docs/llms/openai-compat.md

@@ -8,7 +8,8 @@ nav_order: 500
 Aider can connect to any LLM which is accessible via an OpenAI compatible API endpoint.
 
 ```
-python -m pip install -U aider-chat
+python -m pip install aider-install
+aider-install
 
 # Mac/Linux:
 export OPENAI_API_BASE=<endpoint>