v4.0.0 Live Chroma 4.0 Native Vector Indexing & HNSW Optimizations Released. Read Docs
Cloud Waitlist
Open-Source AI Embedding Engine

The AI-Native Vector Database for Scalable Embeddings.

Chroma gives LLMs memory with ultra-low latency vector search, automatic metadata filtering, and native integration into Python & JS ecosystems.

Get Started Free
$ pip install chromadb
Apache 2.0 Open Source
Sub-5ms Query Latency
embedding_inspector.py
HNSW Index
# Initialize Chroma vector store client
import chromadb
client = chromadb.Client()
collection = client.create_collection(name="docs")
# Query with automatic vector embeddings
results = collection.query(
query_texts=["How to store AI vectors?"],
n_results=2
)
QUERY MATCH RESULT 0.0034s
doc_id: "vec_8829" Distance: 0.082 (Cosine)
doc_id: "vec_1102" Distance: 0.114 (Cosine)

POWERING VECTOR MEMORY FOR HIGH-THROUGHPUT AI TEAMS

ANTHROPIC
COHERE
REPLIT
LANGCHAIN
LLAMA_INDEX
The AI Memory Bottleneck

Traditional Databases Weren't Built for 1536-Dimensional Math.

Relational and keyword-based search systems struggle with high-dimensional embedding lookups. As RAG pipelines scale, queries slow down, cloud costs skyrocket, and setup complexity stalls development.

Query Latency Spikes

Standard SQL database vector extensions hit indexing bottlenecks when dataset sizes cross millions of embeddings.

Complex Infrastructure

Setting up multi-node distributed vector clusters just to prototype a RAG application wastes critical developer hours.

Embedding Disconnect

Manually transforming text to vectors via separate APIs introduces sync bugs, schema drift, and payload bloat.

Engine Architecture

Engineered for Developer Velocity

Simple enough to run in a Jupyter notebook with `import chromadb`. Powerful enough to scale to billions of vectors in production.

Sub-5ms Nearest Neighbor

Optimized C++ core utilizing HNSW indexing algorithms delivers ultra-fast cosine, L2, and inner-product distance calculations.

Rich Metadata Filtering

Combine vector similarity queries with complex boolean metadata filters (`where={"author": "alex", "year": {"$gt": 2024}}`).

Automatic Embedding Functions

Plug-and-play embedding integrations for OpenAI, Cohere, HuggingFace, and local Ollama models with zero boilerplate code.

In-Memory to Serverless

Start completely in-memory for testing, persist to local disk with DuckDB/SQLite, or deploy to distributed Chroma Cloud seamlessly.

Enterprise Security

SOC2 Type II certified infrastructure with full end-to-end encryption at rest and in transit, API key auth, and tenant isolation.

Real-time Telemetry

Built-in Prometheus metrics and OpenTelemetry instrumentation to monitor indexing throughput, memory usage, and latency distribution.

Developer First API

Four Lines of Code to Production Vector Search.

Chroma handles document parsing, embedding generation, indexing, and vector similarity calculations out of the box with an intuitive, clean SDK.

quickstart.py
# 1. Initialize local persistent database
import chromadb
client = chromadb.PersistentClient(path="./chroma_db")

# 2. Get or create collection with metadata
collection = client.get_or_create_collection(
    name="ai_knowledge_base",
    metadata={"hnsw:space": "cosine"}
)

# 3. Add documents with automatic embedding generation
collection.add(
    documents=[
"Chroma v4 provides sub-5ms vector queries",
"Vector databases store high-dimensional embeddings"
    ],
    metadatas=[{"cat": "perf"}, {"cat": "tech"}],
    ids=["id1", "id2"]
)

# 4. Perform vector similarity query
results = collection.query(
    query_texts=["How fast is vector search?"],
    n_results=1
)
                        
Natively Connected

Integrates with Your Favorite AI Stack

Zero-friction connectors for leading model providers and RAG frameworks.

OpenAI Embeddings

text-embedding-3-large & small

LangChain

First-class vector store support

LlamaIndex

Native index storage connector

Hugging Face

Local open-weight embeddings

15M+
Monthly PyPI Downloads
< 4ms
Average Query Latency
10B+
Vectors Indexed Daily
100%
Apache 2.0 Open Source
Predictable Scale

Start Self-Hosted. Scale on Cloud.

Open source is free forever. Chroma Cloud provides managed elasticity.

Community

Open Source

$0 / forever

Full features for local development, research, and self-hosted deployments.

  • Unlimited Local Embeddings
  • In-Memory & Disk Persistence
  • Full HNSW Indexing Support
  • Python & JavaScript SDKs
Install Open Source
Most Popular
Managed Serverless

Chroma Cloud Pro

$29 / month

Fully managed serverless vector store with auto-scaling and zero ops.

  • Up to 10,000,000 Vectors Included
  • 99.9% Uptime SLA
  • Auto Index Optimization
  • Automated Daily Backups
Join Managed Waitlist
High Scale

Enterprise Dedicated

Custom

Dedicated single-tenant clusters in your VPC or ours for high-volume apps.

  • Billion+ Vector Scaling
  • VPC Peering & SOC2 Compliance
  • 24/7 Dedicated Support & SLA
  • Custom Embedding Hardware Accelerator
Contact Architect Team
Developer Feedback

Loved by AI Developers Worldwide

"Chroma completely removed the friction of setting up a vector database for our RAG agent. We went from prototype to production in two days."

AK
Alex Rivera
Lead AI Architect @ Synthetic Inc

"The metadata filtering capabilities in Chroma v4 are ridiculously fast. Filtering millions of customer context vectors takes under 4 milliseconds."

MS
Elena Rostova
Staff ML Engineer @ DataMesh

"Being able to start locally in a Python notebook and migrate seamlessly to cloud infrastructure without code changes is a game changer."

JH
Jason Huang
Founder @ VectorFlow AI
Developer FAQ

Frequently Asked Questions

Ready for Production

Build Smarter RAG Applications Today.

Join tens of thousands of developers using Chroma to store, search, and retrieve high-dimensional AI vectors with zero friction.