Database API Key

ChromaDB REST API

Open-source AI-native embedding database for LLM apps

ChromaDB is an open-source vector database designed for AI applications that need to store and retrieve embeddings efficiently. It provides semantic search capabilities, integrates seamlessly with LLMs, and offers a simple API for managing collections of embeddings with metadata filtering. Developers use ChromaDB to build RAG systems, semantic search engines, and AI-powered applications that require fast similarity search over high-dimensional vectors.

Base URL https://api.chromadb.com/v1

API Endpoints

Method	Endpoint	Description
GET	`/heartbeat`	Check if the ChromaDB server is running and responsive
POST	`/collections`	Create a new collection for storing embeddings with specified metadata
GET	`/collections`	List all collections in the database with their configurations
GET	`/collections/{collection_id}`	Get details about a specific collection including count and metadata
DELETE	`/collections/{collection_id}`	Delete a collection and all its embeddings permanently
POST	`/collections/{collection_id}/add`	Add embeddings, documents, and metadata to a collection
POST	`/collections/{collection_id}/update`	Update existing embeddings, documents, or metadata in a collection
POST	`/collections/{collection_id}/upsert`	Insert or update embeddings based on IDs in a collection
POST	`/collections/{collection_id}/query`	Perform similarity search using query embeddings with filters
POST	`/collections/{collection_id}/get`	Retrieve embeddings by IDs or filter criteria from a collection
POST	`/collections/{collection_id}/delete`	Delete specific embeddings from a collection by IDs or filters
POST	`/collections/{collection_id}/count`	Get the count of embeddings in a collection matching filters
GET	`/version`	Get the ChromaDB server version information
POST	`/tenants`	Create a new tenant for multi-tenancy support
GET	`/tenants`	List all tenants in the ChromaDB instance

Code Examples

# Create a collection
curl -X POST https://api.chromadb.com/v1/collections \
  -H 'Authorization: Bearer YOUR_API_KEY' \
  -H 'Content-Type: application/json' \
  -d '{
    "name": "document_embeddings",
    "metadata": {"description": "Product documentation embeddings"},
    "embedding_function": "default"
  }'

# Add embeddings to collection
curl -X POST https://api.chromadb.com/v1/collections/doc_collection_id/add \
  -H 'Authorization: Bearer YOUR_API_KEY' \
  -H 'Content-Type: application/json' \
  -d '{
    "embeddings": [[0.1, 0.2, 0.3], [0.4, 0.5, 0.6]],
    "documents": ["First document text", "Second document text"],
    "metadatas": [{"source": "docs"}, {"source": "blog"}],
    "ids": ["doc1", "doc2"]
  }'

# Query for similar embeddings
curl -X POST https://api.chromadb.com/v1/collections/doc_collection_id/query \
  -H 'Authorization: Bearer YOUR_API_KEY' \
  -H 'Content-Type: application/json' \
  -d '{
    "query_embeddings": [[0.15, 0.25, 0.35]],
    "n_results": 5,
    "where": {"source": "docs"}
  }'

// Create a collection
const createCollection = async () => {
  const response = await fetch('https://api.chromadb.com/v1/collections', {
    method: 'POST',
    headers: {
      'Authorization': 'Bearer YOUR_API_KEY',
      'Content-Type': 'application/json'
    },
    body: JSON.stringify({
      name: 'document_embeddings',
      metadata: { description: 'Product documentation embeddings' },
      embedding_function: 'default'
    })
  });
  return await response.json();
};

// Add embeddings
const addEmbeddings = async (collectionId) => {
  const response = await fetch(`https://api.chromadb.com/v1/collections/${collectionId}/add`, {
    method: 'POST',
    headers: {
      'Authorization': 'Bearer YOUR_API_KEY',
      'Content-Type': 'application/json'
    },
    body: JSON.stringify({
      embeddings: [[0.1, 0.2, 0.3], [0.4, 0.5, 0.6]],
      documents: ['First document text', 'Second document text'],
      metadatas: [{ source: 'docs' }, { source: 'blog' }],
      ids: ['doc1', 'doc2']
    })
  });
  return await response.json();
};

// Query similar embeddings
const querySimilar = async (collectionId) => {
  const response = await fetch(`https://api.chromadb.com/v1/collections/${collectionId}/query`, {
    method: 'POST',
    headers: {
      'Authorization': 'Bearer YOUR_API_KEY',
      'Content-Type': 'application/json'
    },
    body: JSON.stringify({
      query_embeddings: [[0.15, 0.25, 0.35]],
      n_results: 5,
      where: { source: 'docs' }
    })
  });
  return await response.json();
};

import requests
import json

API_KEY = 'YOUR_API_KEY'
BASE_URL = 'https://api.chromadb.com/v1'
HEADERS = {
    'Authorization': f'Bearer {API_KEY}',
    'Content-Type': 'application/json'
}

# Create a collection
def create_collection():
    response = requests.post(
        f'{BASE_URL}/collections',
        headers=HEADERS,
        json={
            'name': 'document_embeddings',
            'metadata': {'description': 'Product documentation embeddings'},
            'embedding_function': 'default'
        }
    )
    return response.json()

# Add embeddings to collection
def add_embeddings(collection_id):
    response = requests.post(
        f'{BASE_URL}/collections/{collection_id}/add',
        headers=HEADERS,
        json={
            'embeddings': [[0.1, 0.2, 0.3], [0.4, 0.5, 0.6]],
            'documents': ['First document text', 'Second document text'],
            'metadatas': [{'source': 'docs'}, {'source': 'blog'}],
            'ids': ['doc1', 'doc2']
        }
    )
    return response.json()

# Query for similar embeddings
def query_similar(collection_id):
    response = requests.post(
        f'{BASE_URL}/collections/{collection_id}/query',
        headers=HEADERS,
        json={
            'query_embeddings': [[0.15, 0.25, 0.35]],
            'n_results': 5,
            'where': {'source': 'docs'}
        }
    )
    return response.json()

# Get collection details
def get_collection(collection_id):
    response = requests.get(
        f'{BASE_URL}/collections/{collection_id}',
        headers=HEADERS
    )
    return response.json()

Use ChromaDB from Claude / Cursor / ChatGPT

Get a hosted MCP endpoint for ChromaDB. Paste your ChromaDB API key, copy back one URL, drop it into Claude Desktop, Cursor, or any AI client that supports remote MCP. Your AI calls ChromaDB directly with your credentials — no local install, works on mobile.

chromadb_create_collection Create a new ChromaDB collection with specified name and metadata for organizing embeddings

chromadb_add_embeddings Add document embeddings with metadata to a collection for semantic search and retrieval

chromadb_semantic_search Perform similarity search across embeddings using query vectors with optional metadata filters

chromadb_retrieve_documents Retrieve documents and embeddings by IDs or metadata filters from a collection

chromadb_delete_embeddings Delete specific embeddings from a collection by IDs or matching metadata criteria

Connect in 60 seconds

Paste your ChromaDB key → get an MCP URL → paste into Claude/Cursor. Hosted by IOX, encrypted at rest.

Connect ChromaDB to your AI →

ChromaDB REST API

API Endpoints

Sponsor this page

Code Examples

Use ChromaDB from Claude / Cursor / ChatGPT

Connect in 60 seconds

Related APIs