#1313th of 26 in RAG and knowledge
docsgpt
Builds agents and workflows that answer from your documents with citations
Live demo ↗
(opens in a new tab)Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 18.3k
- License
- MIT
- Last commit
- Oct 2026
- Last release
- Oct 2026
- Language
- Python
Missed the latest GitHub check; these facts may be older.
Overview
DocsGPT ingests documents, web pages and synced sources such as Google Drive, Confluence, GitHub and S3, and runs agents that answer with cited sources. Agents can use tools, MCP servers and sandboxed code, and can be chained in a visual workflow builder. It is served through a web app, an embeddable widget, an OpenAI-compatible /v1 API and an MCP server, with OIDC SSO, teams and quotas.
Who it is for: Teams self-hosting document-grounded agents and support bots
Strengths
- Connectors sync Google Drive, SharePoint, Confluence, GitHub and S3, with GraphRAG
- Exposes agents via widget, OpenAI-compatible API, webhooks and MCP server
- SSO (OIDC/SCIM), roles, quotas and audit log are all MIT licensed
- Works with OpenAI, Anthropic, Google, Groq, Ollama, vLLM; air-gapped install documented
Weaknesses
- Stack needs API, Celery worker, Postgres, Redis and a vector store
- README gives no RAM or GPU sizing for local models
- Wide scope (workflows, guardrails, connectors) means a large surface to operate
- Quick install is a curl-pipe-to-shell script that requires Docker
What it needs
- GPU optional
- Docker + Compose
- Needs Docker, PostgreSQL, Redis, Vector store (FAISS, pgvector, Elasticsearch, Qdrant, Milvus or MongoDB)
- Models: OpenAI, Anthropic, Google, Groq, OpenRouter
- port 7091
Also in RAG and knowledge
See all 26| Rank | Project | Score |
|---|---|---|
| 1 | deeptutorSelf-hosted AI tutor with knowledge bases, guided learning, books and quizzes | 80 out of 100 |
| 2 | Open NotebookSelf-hosted NotebookLM alternative with podcasts and 20+ model providers | 76 out of 100 |
| 3 | LightRAGGraph-based RAG framework with a REST API server and web UI | 76 out of 100 |
| #4 | RAGFlowRAG engine with deep document parsing, template chunking and agentic retrieval | 74 out of 100 |
| #5 | MaxKBEnterprise knowledge-base agent platform with RAG, workflows and MCP tools | 71 out of 100 |