Products & Announcements

Onyx: Open-source enterprise chat UI for any LLM with RAG, tools, and deep research

Open SourceRetrieval-Augmented GenerationCorporate AI StrategySelf-HostingLLM Context Management
Added Nov 25, 2025
Link160
Article: PositiveCommunity: PositiveDivisive

Onyx is an open-source, model-agnostic chat UI that brings enterprise features and integrated tools like RAG, web search, MCP, deep research, and a secure code interpreter. Born from Danswer, it focuses on consumer-grade UX for everyday work while supporting complex agentic workflows and on-prem, air-gapped deployments. The team highlights critical engineering tactics for context management and model-specific tool behavior, and reports early traction with large enterprises.

Key Points

  • Pivot from Danswer: Users primarily wanted a secure, high-quality chat UI over full enterprise search, prompting the creation of Onyx.
  • Focus on UX: Matching and exceeding consumer-grade chat experiences for work use is hard but essential for adoption.
  • Full toolchain integration: RAG, web search, MCP, deep research, memory, assistants, and a secure, model-agnostic code interpreter.
  • Enterprise-ready: RBAC, SSO, permission syncing, and easy on-prem/air-gapped hosting for sensitive industries.
  • Engineering learnings: Use end-of-message “Reminder” prompts for instruction adherence and adapt tool behavior to model tendencies.

Sentiment

The community is cautiously receptive. Enterprise users and open-source advocates see genuine value in model-agnostic chat with strong RAG, but persistent skepticism about differentiation in a crowded market, bloated deployment requirements, and the sustainability of the open-core business model prevent full enthusiasm. The founders' extensive engagement in the thread is noted positively.

In Agreement

  • Enterprise need for model-agnostic, self-hostable LLM chat is real and validated by Fortune 100 users struggling with Copilot and failed internal builds
  • The RAG and connector ecosystem with community-contributed integrations provides genuine value for unified enterprise search across disparate systems
  • Open-source customizability and white-labeling fill a gap that ChatGPT Enterprise and Copilot cannot address
  • Model flexibility matters as leading models constantly change across providers, making vendor lock-in costly
  • The founders' enterprise search background gives them a meaningful advantage in RAG quality over general chat alternatives

Opposed

  • The market is extremely crowded with similar alternatives like OpenWebUI, LibreChat, and AnythingLLM, and there is no clear moat
  • The Docker installation is bloated and resource-heavy compared to lighter alternatives, making it inaccessible for personal or small-scale deployments
  • Calling the project 'open source' while gating enterprise features behind proprietary licenses is misleading open-core at best
  • VC funding creates misaligned incentives that will eventually lead to enshittification or mission drift away from the open-source community
  • The product feels underbaked in key areas like admin UX, document management ergonomics, and document processing workflows
  • A general-purpose 'one chat to rule them all' approach may lose to specialized vertical solutions tailored to specific industries
Onyx: Open-source enterprise chat UI for any LLM with RAG, tools, and deep research | TD Stuff