Trieve Documentation

Trieve is the best GenAI platform for creating, managing, and searching embeddings then using them via Retrieval-Augmented Generation (RAG) or your method of choice.

Build semantic search and RAG fast.

Things still unclear? Check out our live demos or REST OpenAPI docs

We have deployed Trieve for production datasets of the Enron email corpus and for YCombinator companies.

Getting started

If you are interested in a managed instance and/or SLA, reach out to us by emailing

To engage with the Trieve community, preferably join our Matrix, but we also have a Discord.



Learn how to quickly get started with Trieve.

Read more

Self Hosting

Learn how to self-host Trieve on your own infrastructure.

Read more

Environment Variables

Learn about the different environment variables used by Trieve.

Read more

Use Custom LLM or Embedding Model

Learn how to use custom embedding models and LLMs with Trieve.

Read more



Learn about how to authenticate with the Trieve API and how to use the access cookie and API key to make requests.


Learn about the chunk model and how to create, retrieve, update, delete, and query chunks.


Learn about the collection model and how to create, retrieve, update, delete, and list collections.


Learn about the user model and how to create, retrieve, update, and delete users and their associated info.


Learn about the file model and how to upload, retrieve, update, delete, and list files.