onyx-dot-app/onyx-foss

Open Source AI Platform - AI Chat with advanced features that works with every LLM

What it solves

Onyx provides a feature-rich application layer for Large Language Models (LLMs), allowing users and organizations to host their own AI platform. It bridges the gap between raw LLMs and usable business tools by integrating RAG, web search, and external application connectivity into a single interface.

How it works

Onyx acts as an orchestrator that connects to various LLM providers (both proprietary like OpenAI and self-hosted like Ollama). It uses a hybrid index (vector and keyword) for Agentic RAG and provides a sandboxed environment for code execution. The platform supports two deployment modes: a "Lite" version for basic chat and agents, and a "Standard" version that includes full indexing, job queues for knowledge syncing, and performance optimizations via Redis and MinIO.

Who it’s for

It is designed for a wide range of users, from individuals wanting a self-hosted AI interface to large enterprises requiring collaboration tools, Single Sign-On (SSO), Role-Based Access Control (RBAC), and whitelabeling capabilities.

Highlights

  • Agentic RAG: Combines hybrid indexing with AI agents for high-quality information retrieval.
  • Deep Research: A multi-step research flow for generating in-depth reports.
  • Extensive Connectivity: Over 50 indexing-based connectors and support for the Model Context Protocol (MCP).
  • Custom Agents: Ability to build agents with specific instructions, knowledge bases, and actions.
  • Versatile Tooling: Built-in web crawling, image generation, voice mode (TTS/STT), and sandboxed code execution for data analysis.

Related

  • Project
    onyx-dot-app/onyx

    Onyx is an open-source AI platform that serves as an application layer for LLMs, providing advanced capabilities like Agentic RAG, deep research, and custom agents in a self-hostable interface.

  • Project
    brainlid/langchain

    An Elixir framework designed to help developers build data-aware and agentic applications by chaining LLMs with external tools, data, and services.

  • Project
    ogx-ai/ogx

    OGX is an open‑source, OpenAI‑compatible API server that lets you run any LLM (Llama, GPT, Gemini, Mistral, etc.) on any infrastructure. It provides standard chat/completions, embeddings, vector stores, files, batches, skills, and an agentic Responses API, while also supporting Anthropic and Google GenAI SDKs out of the box.

  • Project
    open-webui/open-webui

    A feature-rich, self-hosted AI platform that provides a user-friendly interface for local and API-based LLMs, featuring built-in RAG and extensive plugin support.

  • Project
    the-open-agent/openagent

    An open-source personal AI assistant platform that combines LLMs, RAG, and autonomous agent loops into a single, self-hostable binary.