mirror of
https://github.com/pykeio/ort
synced 2026-04-25 16:34:55 +02:00
39 lines
2.7 KiB
Plaintext
39 lines
2.7 KiB
Plaintext
---
|
|
title: Cargo features
|
|
---
|
|
|
|
# Cargo features
|
|
|
|
> *✅ = default, ⚒️ = not default*
|
|
|
|
- ✅ **`ndarray`**: Enables tensors to be created from/extracted to [`ndarray`](https://crates.io/crates/ndarray) multi-dimensional arrays. We highly recommend this feature if you need to do a lot of complex pre/post-processing requiring multi-dimensional array access, but for something like an LLM, omitting this feature won't require too much extra work but will save a fair amount of compile time.
|
|
- ✅ **`download-binaries`**: Downloads prebuilt binaries from pyke's CDN service for supported platforms. Disabling this means you'll need to compile ONNX Runtime from source yourself, and [link `ort` to it](/setup/linking).
|
|
- ✅ **`copy-dylibs`**: In case dynamic libraries are used (like with the CUDA execution provider), creates a symlink to them in the relevant places in the `target` folder to make [compile-time dynamic linking](/setup/linking#compile-time-dynamic-linking) work.
|
|
- ⚒️ **`half`**: Enables support for creating & extracting float16/bfloat16 tensors via the [`half`](https://crates.io/crates/half) crate. ONNX models that are converted to 16-bit precision will typically convert to/from 32-bit floats at the input/output, so you will likely never actually need to interact with a 16-bit tensor on the Rust side.
|
|
- ⚒️ **`num-complex`**: Enables support for creating & extracting complex32/complex64 tensors via the [`num-complex`](https://crates.io/crates/num-complex) crate.
|
|
- ⚒️ **`load-dynamic`**: Enables [runtime dynamic linking](/setup/linking#runtime-loading-with-load-dynamic), which alleviates many of the troubles with compile-time dynamic linking and offers greater flexibility.
|
|
- ⚒️ **`alternative-backend`**: Disables linking to ONNX Runtime, allowing you to instead configure an [alternative backend](/backends).
|
|
- ⚒️ **`fetch-models`**: Enables the [`SessionBuilder::commit_from_url`](https://docs.rs/ort/2.0.0-rc.9/ort/session/builder/struct.SessionBuilder.html#method.commit_from_url) method, allowing you to quickly download & run a model from a URL. This should only be used for quick testing.
|
|
|
|
## Execution providers
|
|
Each [execution provider](/perf/execution-providers) is also gated behind a Cargo feature.
|
|
- **NVIDIA CUDA**: `cuda`
|
|
- **NVIDIA TensorRT**: `tensorrt`
|
|
- **Microsoft DirectML**: `directml`
|
|
- **Apple CoreML**: `coreml`
|
|
- **AMD ROCm**: `rocm`
|
|
- **Intel OpenVINO**: `openvino`
|
|
- **Intel oneDNN**: `onednn`
|
|
- **XNNPACK**: `xnnpack`
|
|
- **Qualcomm QNN**: `qnn`
|
|
- **Huawei CANN**: `cann`
|
|
- **Android NNAPI**: `nnapi`
|
|
- **Apache TVM**: `tvm`
|
|
- **Arm ACL**: `acl`
|
|
- **ArmNN**: `armnn`
|
|
- **AMD MIGraphX**: `migraphx`
|
|
- **AMD Vitis AI**: `vitis`
|
|
- **Rockchip RKNPU**: `rknpu`
|
|
- **WebGPU**: `webgpu`
|
|
- **Microsoft Azure**: `azure`
|