Favicon of Onyx

Onyx

Self-hosted AI search across company apps, with local-model options and source-permission sync in the Enterprise Edition.

Screenshot of Onyx website

Onyx is an AI search and chat platform for teams whose information is spread across workplace apps. It indexes company knowledge so employees can ask questions across sources and get answers grounded in relevant documents. Teams can deploy it in their own cloud or on bare metal, including an air-gapped environment.

Search draws on indexed content from tools such as GitHub, Google Drive, Slack and Confluence. In the Enterprise Edition, Onyx can sync source permissions into its index and check them when someone asks a question, so an answer reflects what that person can access. Employees can use the web or desktop app, while teams can bring the same search into Slack. An MCP server gives coding assistants such as Claude Code and Codex access to company context under those permissions.

Onyx also supports custom AI agents with selected knowledge, instructions and actions. They can use web search and external applications, and the platform can produce reports and downloadable artifacts. For teams choosing their own models, Onyx works with local backends including Ollama, LiteLLM and vLLM, as well as providers such as Anthropic, OpenAI and Gemini. With local models, the document index, embeddings and model traffic can stay inside the team's infrastructure; using an external model provider sends model traffic outside that boundary.

The Community Edition is MIT-licensed and covers core search, chat, agents and actions. Self-hosted Enterprise adds source-permission sync, role controls and SSO. Onyx Cloud is available for teams that do not want to run the platform themselves.

Similar to Onyx