Onyx

Open-source AI chat connected to your docs, apps, and people

Freemium· Self-hosted open-source: free. Business Cloud: $20 per user/month (annual). Enterprise: custom pricing with SSO, on-prem, region deployments, and SLA.RAGLLM-agnostic — routes to OpenAI (GPT-4o/GPT-5), Anthropic Claude, Google Gemini, Azure OpenAI, AWS Bedrock, or local Ollama/vLLM models

Visit website →

Best for

Mid-size and enterprise teams that want a permission-aware, LLM-agnostic RAG chatbot over Google Drive, Slack, Confluence, Jira, and code repos — either self-hosted for data sovereignty or as a managed cloud with SSO.

Skip if

Solo users or small teams who just want a personal ChatGPT wrapper, and shops that need a lightweight no-ops SaaS assistant without running Docker, Postgres, and a vector store.

Onyx (formerly known as Danswer) is an open-source enterprise search and AI chat platform that plugs a unified assistant into a company's internal knowledge base. It ingests content from 40+ connectors — Google Drive, Confluence, Notion, Slack, GitHub, Jira, Salesforce, Zendesk, SharePoint, S3, web pages, and more — and exposes it through a ChatGPT-style UI that grounds answers in retrieved documents with citations back to the source. Under the hood it runs a hybrid retrieval stack (BM25 plus dense embeddings) with a re-ranker, and supports pluggable LLM providers so teams can route to OpenAI, Anthropic, Azure, Bedrock, Vertex, Ollama, or self-hosted open-weight models. Beyond straight Q&A, Onyx ships a custom-agent framework, a code interpreter, image generation, web search, deep-research mode, and a Slack bot that answers threads in-channel. Access control is a first-class feature: connectors respect the source system's permissions so a user only ever sees retrieved snippets they were already authorized to read, which is what makes it viable inside larger organizations. Teams typically deploy it as an internal help-desk brain for support and sales, a code and design docs lookup for engineering, or a company-wide research assistant that replaces scattered knowledge searches. It is available as self-hosted Docker/Kubernetes for full data-plane control, a managed cloud with SOC 2 Type II and GDPR compliance, or an enterprise install with SSO, on-prem, and region-locked deployment.

Editor's take

Onyx is the strongest open-source pick if you need enterprise search that respects source-system permissions, and the LLM-agnostic backend means you are not locked into one provider. Self-hosting is real work — plan for a small ops footprint — but the payoff is a RAG stack you can actually audit, versus a black-box SaaS.

— The AI Tool Bible editorial team

Pros

✅ Open-source (MIT-adjacent) with active development and 20k+ GitHub stars, so you can self-host and audit the retrieval pipeline
✅ 40+ pre-built connectors for common SaaS and file stores, saving weeks of custom ETL work
✅ Permission-aware retrieval that honors source-system ACLs, avoiding the classic RAG leak of exposing restricted docs
✅ LLM-agnostic: swap between GPT, Claude, Gemini, Bedrock, or a local Ollama/vLLM model without rewriting the stack
✅ Hybrid search plus re-ranking out of the box, rather than a naive top-k vector lookup
✅ Custom agent framework, code interpreter, and Slack bot ship in-product, not as separate SKUs
✅ Cloud tier gives a managed option with SOC 2 Type II, GDPR, SSO, and audit logs for enterprise buyers

Cons

⚠️ Self-hosting is Docker/K8s-heavy and needs Postgres, Vespa/Vector store, and worker processes — not a one-click install for small teams
⚠️ Answer quality still depends heavily on your connector hygiene; stale or duplicated source docs produce confidently wrong answers
⚠️ Cloud pricing at $20/user/mo scales quickly for large orgs versus running the OSS build yourself
⚠️ Custom agent authoring is less mature than dedicated agent-builder tools like LangGraph or CrewAI
⚠️ Fine-grained observability (per-query latency, retrieval traces) is thinner than specialist LLMOps platforms

Use cases

Internal knowledge-base chatbot over Confluence and Google DriveSupport-team assistant grounded in Zendesk tickets and help docsSales enablement over Salesforce, Gong, and pitch decksEngineering docs and codebase Q&A over GitHub and NotionSlack bot that answers questions in-thread with citationsDeep-research agent across web and internal sourcesOnboarding assistant for new hiresPermission-scoped RAG for regulated industries