Chroma Cloud vector db provider (#4273)

* add chroma cloud as new vector db provider

* update docker example env

* extend chroma class to chroma cloud

* update readme

---------

Co-authored-by: timothycarambat <rambat1010@gmail.com>

README.md

@@ -136,7 +136,7 @@ AnythingLLM divides your documents into objects called `workspaces`. A Workspace
 - [PGVector](https://github.com/pgvector/pgvector)
 - [Astra DB](https://www.datastax.com/products/datastax-astra)
 - [Pinecone](https://pinecone.io)
-- [Chroma](https://trychroma.com)
+- [Chroma & ChromaCloud](https://trychroma.com)
 - [Weaviate](https://weaviate.io)
 - [Qdrant](https://qdrant.tech)
 - [Milvus](https://milvus.io)

docker/.env.example

@@ -212,6 +212,12 @@ GID='1000'
 # CHROMA_API_HEADER="X-Api-Key"
 # CHROMA_API_KEY="sk-123abc"
 
+# Enable all below if you are using vector database: Chroma Cloud.
+# VECTOR_DB="chromacloud"
+# CHROMACLOUD_API_KEY="ck-your-api-key"
+# CHROMACLOUD_TENANT=
+# CHROMACLOUD_DATABASE=
+
 # Enable all below if you are using vector database: Pinecone.
 # VECTOR_DB="pinecone"
 # PINECONE_API_KEY=

frontend/src/components/VectorDBSelection/ChromaCloudOptions/index.jsx

@@ -0,0 +1,53 @@
+export default function ChromaCloudOptions({ settings }) {
+  return (
+    <div className="w-full flex flex-col gap-y-7">
+      <div className="w-full flex items-center gap-[36px] mt-1.5">
+        <div className="flex flex-col w-60">
+          <label className="text-white text-sm font-semibold block mb-3">
+            API Key
+          </label>
+          <input
+            type="password"
+            name="ChromaCloudApiKey"
+            className="border-none bg-theme-settings-input-bg text-white placeholder:text-theme-settings-input-placeholder text-sm rounded-lg focus:outline-primary-button active:outline-primary-button outline-none block w-full p-2.5"
+            placeholder="ck-your-api-key-here"
+            defaultValue={settings?.ChromaCloudApiKey ? "*".repeat(20) : ""}
+            required={true}
+            autoComplete="off"
+            spellCheck={false}
+          />
+        </div>
+
+        <div className="flex flex-col w-60">
+          <label className="text-white text-sm font-semibold block mb-3">
+            Tenant ID
+          </label>
+          <input
+            name="ChromaCloudTenant"
+            autoComplete="off"
+            type="text"
+            defaultValue={settings?.ChromaCloudTenant}
+            className="border-none bg-theme-settings-input-bg text-white placeholder:text-theme-settings-input-placeholder text-sm rounded-lg focus:outline-primary-button active:outline-primary-button outline-none block w-full p-2.5"
+            placeholder="your-tenant-id-here"
+            required={true}
+          />
+        </div>
+
+        <div className="flex flex-col w-60">
+          <label className="text-white text-sm font-semibold block mb-3">
+            Database Name
+          </label>
+          <input
+            name="ChromaCloudDatabase"
+            autoComplete="off"
+            type="text"
+            defaultValue={settings?.ChromaCloudDatabase}
+            className="border-none bg-theme-settings-input-bg text-white placeholder:text-theme-settings-input-placeholder text-sm rounded-lg focus:outline-primary-button active:outline-primary-button outline-none block w-full p-2.5"
+            placeholder="your-database-name"
+            required={true}
+          />
+        </div>
+      </div>
+    </div>
+  );
+}

frontend/src/pages/GeneralSettings/VectorDatabase/index.jsx

@@ -24,6 +24,7 @@ import PGVectorLogo from "@/media/vectordbs/pgvector.png";
 
 import LanceDBOptions from "@/components/VectorDBSelection/LanceDBOptions";
 import ChromaDBOptions from "@/components/VectorDBSelection/ChromaDBOptions";
+import ChromaCloudOptions from "@/components/VectorDBSelection/ChromaCloudOptions";
 import PineconeDBOptions from "@/components/VectorDBSelection/PineconeDBOptions";
 import WeaviateDBOptions from "@/components/VectorDBSelection/WeaviateDBOptions";
 import QDrantDBOptions from "@/components/VectorDBSelection/QDrantDBOptions";
@@ -133,6 +134,14 @@ export default function GeneralVectorDatabase() {
       description:
         "Open source vector database you can host yourself or on the cloud.",
     },
+    {
+      name: "Chroma Cloud",
+      value: "chromacloud",
+      logo: ChromaLogo,
+      options: <ChromaCloudOptions settings={settings} />,
+      description:
+        "Fully managed Chroma cloud service with enterprise features and support.",
+    },
     {
       name: "Pinecone",
       value: "pinecone",

frontend/src/pages/OnboardingFlow/Steps/DataHandling/index.jsx

@@ -271,6 +271,14 @@ export const VECTOR_DB_PRIVACY = {
     ],
     logo: ChromaLogo,
   },
+  chromacloud: {
+    name: "Chroma Cloud",
+    description: [
+      "Your vectors and document text are stored on Chroma's cloud service",
+      "Access to your data is managed by Chroma",
+    ],
+    logo: ChromaLogo,
+  },
   pinecone: {
     name: "Pinecone",
     description: [

server/.env.example

@@ -202,6 +202,12 @@ SIG_SALT='salt' # Please generate random string at least 32 chars long.
 # CHROMA_API_HEADER="X-Api-Key"
 # CHROMA_API_KEY="sk-123abc"
 
+# Enable all below if you are using vector database: Chroma Cloud.
+# VECTOR_DB="chromacloud"
+# CHROMACLOUD_API_KEY="ck-your-api-key"
+# CHROMACLOUD_TENANT=
+# CHROMACLOUD_DATABASE=
+
 # Enable all below if you are using vector database: Pinecone.
 # VECTOR_DB="pinecone"
 # PINECONE_API_KEY=

server/models/systemSettings.js

@@ -428,6 +428,11 @@ const SystemSettings = {
       ChromaApiHeader: process.env.CHROMA_API_HEADER,
       ChromaApiKey: !!process.env.CHROMA_API_KEY,
 
+      // ChromaCloud DB Keys
+      ChromaCloudApiKey: !!process.env.CHROMACLOUD_API_KEY,
+      ChromaCloudTenant: process.env.CHROMACLOUD_TENANT,
+      ChromaCloudDatabase: process.env.CHROMACLOUD_DATABASE,
+
       // Weaviate DB Keys
       WeaviateEndpoint: process.env.WEAVIATE_ENDPOINT,
       WeaviateApiKey: process.env.WEAVIATE_API_KEY,

server/utils/helpers/index.js

@@ -77,7 +77,7 @@
 
 /**
  * Gets the systems current vector database provider.
- * @param {('pinecone' | 'chroma' | 'lancedb' | 'weaviate' | 'qdrant' | 'milvus' | 'zilliz' | 'astra') | null} getExactly - If provided, this will return an explit provider.
+ * @param {('pinecone' | 'chroma' | 'chromacloud' | 'lancedb' | 'weaviate' | 'qdrant' | 'milvus' | 'zilliz' | 'astra') | null} getExactly - If provided, this will return an explit provider.
  * @returns { BaseVectorDatabaseProvider}
  */
 function getVectorDbClass(getExactly = null) {
@@ -89,6 +89,9 @@ function getVectorDbClass(getExactly = null) {
     case "chroma":
       const { Chroma } = require("../vectorDbProviders/chroma");
       return Chroma;
+    case "chromacloud":
+      const { ChromaCloud } = require("../vectorDbProviders/chromacloud");
+      return ChromaCloud;
     case "lancedb":
       const { LanceDb } = require("../vectorDbProviders/lance");
       return LanceDb;

server/utils/helpers/updateENV.js

@@ -332,6 +332,20 @@ const KEY_MAPPING = {
     checks: [],
   },
 
+  // ChromaCloud Options
+  ChromaCloudApiKey: {
+    envKey: "CHROMACLOUD_API_KEY",
+    checks: [isNotEmpty],
+  },
+  ChromaCloudTenant: {
+    envKey: "CHROMACLOUD_TENANT",
+    checks: [isNotEmpty],
+  },
+  ChromaCloudDatabase: {
+    envKey: "CHROMACLOUD_DATABASE",
+    checks: [isNotEmpty],
+  },
+
   // Weaviate Options
   WeaviateEndpoint: {
     envKey: "WEAVIATE_ENDPOINT",
@@ -845,6 +859,7 @@ function supportedEmbeddingModel(input = "") {
 function supportedVectorDB(input = "") {
   const supported = [
     "chroma",
+    "chromacloud",
     "pinecone",
     "lancedb",
     "weaviate",

server/utils/vectorDbProviders/chromacloud/index.js

@@ -0,0 +1,29 @@
+const { CloudClient } = require("chromadb");
+const { Chroma } = require("../chroma");
+
+// ChromaCloud works exactly the same as Chroma so we can just extend the
+// Chroma class and override the connect method to use CloudClient
+
+const ChromaCloud = {
+  ...Chroma,
+  name: "ChromaCloud",
+  connect: async function () {
+    if (process.env.VECTOR_DB !== "chromacloud")
+      throw new Error("ChromaCloud::Invalid ENV settings");
+
+    const client = new CloudClient({
+      apiKey: process.env.CHROMACLOUD_API_KEY,
+      tenant: process.env.CHROMACLOUD_TENANT,
+      database: process.env.CHROMACLOUD_DATABASE,
+    });
+
+    const isAlive = await client.heartbeat();
+    if (!isAlive)
+      throw new Error(
+        "ChromaCloud::Invalid Heartbeat received - is the instance online?"
+      );
+    return { client };
+  },
+};
+
+module.exports.ChromaCloud = ChromaCloud;