Chroma is an AI-native open-source vector database. It comes with everything you need to get started built in, and runs on your machine. A hosted version is in early access!

1. Install#

2. Create a Chroma Client#

3. Create a collection#

Collections are where you'll store your embeddings, documents, and any additional metadata. You can create a collection with a name:

4. Add some text documents to the collection#

Chroma will store your text and handle embedding and indexing automatically. You can also customize the embedding model.

5. Query the collection#

You can query the collection with a list of query texts, and Chroma will return the n most similar results. It's that easy!

6. Inspect Results#

From the above query - you can see that our query about hawaii is the semantically most similar to the document about pineapple. This, intuitively, makes sense!

7. Try it out yourself#

For example - what if we tried querying with "This is a document about florida"?

📚 Next steps#

• Read the 🧪 Usage Guide to learn more about the API
• Learn how to ☁️ Deploy Chroma to a server
• Join Chroma's Discord Community to ask questions and get help
• Follow Chroma on Twitter (@trychroma) for updates