AIAny
AI Client2023
Icon for item

Onyx

Self-hostable chat UI that connects to any LLM and adds Agents, Web Search, RAG, connectors, code execution and image generation. Ships connectors to 40+ sources and deployment guides for Docker/K8s. Best for teams needing private, extensible chat platforms.

Introduction

Most teams building RAG or agent workflows still stitch together a UI, retriever, connectors, and orchestration. That integration friction is the problem Onyx targets: it bundles a production-ready chat interface with multi-step agents, hybrid search, rich connectors, and deployment options so organizations can ship a managed conversational AI experience without rebuilding orchestration.

What Sets It Apart
  • Built-in agent orchestration and multi-step "Deep Research" flows — so you get agentic workflows (search → synthesize → act) out of the box rather than wiring separate orchestrators.
  • Hybrid RAG + knowledge graph retrieval with 40+ connectors — this makes ingesting enterprise sources (Google Drive, Slack, databases, etc.) and keeping permissions synchronized significantly easier for private deployments.
  • Deploy-first design (Docker, Kubernetes, Terraform) plus an optional cloud offering — meaning teams can run fully air-gapped installs or scale to multi-node enterprise setups with RBAC, SSO and credential encryption.
  • Provider-agnostic LLM support — works with hosted APIs (OpenAI, Anthropic, Gemini...) or self-hosted models (Ollama, vLLM), so you can trade latency/cost/privacy according to constraints.
Who It's For & Trade-offs

Great fit if you: need an enterprise-grade chat surface that enforces document permissions and integrates many knowledge sources; want to self-host or run in an air-gapped environment; require agent workflows and code execution inside chats. Look elsewhere if you: only need a lightweight single-user chat client (Onyx is feature-rich and operationally heavier), or if you need a fully managed hosted LLM + UI with no infra responsibility at all.

Where It Fits

Onyx sits between low-level retrieval/orchestration libraries and closed hosted chat products: it reduces assembly cost for teams that want control over data, connectors, and deployment while retaining flexibility to swap underlying LLM providers. Use it when integration and permissions matter more than having the absolute simplest UI.

Information

  • Websitegithub.com
  • Authorsonyx-dot-app
  • Published date2023/04/27

More Items

GitHub
AI Deploy2026

Provides reusable Docker compose files, scripts and benchmarked configs to serve modern LLMs (Qwen, Gemma, etc.) on 1–2 NVIDIA RTX 3090/4090/5090 GPUs. Multi-engine (vLLM, llama.cpp, ik_llama), measured TPS/context tradeoffs, and validated single/dual‑GPU recipes.

GitHub
AI Client2025

Enhances web AI chat interfaces (Gemini, AI Studio, Claude, ChatGPT) with folder-based conversation organization, timeline navigation, prompt vault, cloud sync, chat export, UI fixes and a plugin engine. Runs as a browser extension and includes cross-site prompt management and Mermaid rendering.

GitHub
AI Deploy2023

Provides a unified platform for AI development and deployment, including the MAX serving framework and the Mojo systems programming language for writing kernels. Offers an OpenAI-compatible inference endpoint, Mojo-written CPU/GPU kernels, and tools to run hundreds of open models across diverse hardware without vendor lock-in.