
AnythingLLM
6.8/10Open-source app from Mintplex Labs for chatting with your own documents and running AI agents, with any local or cloud LLM, on desktop, self-hosted, or in the cloud.
Data last updated November 2026 · Our methodology
Overview
AnythingLLM is an MIT-licensed application from Mintplex Labs that combines document ingestion, a vector database and connections to large language models in one interface. Teams create workspaces. Each workspace keeps its own documents, chat history and model settings, and they can be pointed at local runtimes such as Ollama or LM Studio, or at cloud providers such as OpenAI, Anthropic, Azure OpenAI, Gemini, Groq and Mistral. LanceDB is built in as the default vector store, and Pinecone, Chroma, Qdrant, Milvus and Weaviate are also supported. An @agent mode and Agent Flows handle multi-step tasks such as web search and calls to external APIs. The documentation also lists MCP compatibility, scheduled jobs, text-to-speech and speech-to-text models, an embeddable chat widget and a developer API. It suits privacy-conscious small teams, developers and IT departments that want documents to stay on their own hardware. The desktop app and the Docker self-hosted version are free. Hosted cloud plans cost $50 a month for Basic and $99 a month for Pro, and on both you bring your own LLM API key and pay the provider separately. Enterprise pricing is by quote and adds on-premise deployment, SSO, RBAC, custom integrations and a custom SLA. The main limitation is shallow RAG tuning. Reviewers say chunking and retrieval offer little control and are hard to debug, and agent orchestration is basic next to dedicated frameworks. The desktop app supports only one user, so shared use requires Docker or the cloud.
Our Rating
Who AnythingLLM is best for
Best software for DevelopersGood fit
Open source and self-hostable, with an API, MCP compatibility and support for many LLMs and vector databases.
IT can run it on-premise with Docker and user permissions so documents never leave company infrastructure.
The free self-hosted version and the $50 cloud plan make private document chat affordable.
Less suited
Local, offline use keeps sensitive files private, but retrieval tuning is limited.
SSO and RBAC exist only on the quote-based Enterprise plan, and reviewers note audit and permission controls are limited at scale.
Pros & Cons
Pros
- The desktop and Docker versions are free under the MIT license, with no feature paywall.
- It can run fully offline with a local model and the built-in LanceDB, so documents never leave your machine.
- You can change the LLM or vector database for each workspace without rebuilding the workspace.
- One reviewer says a working setup takes about 15 minutes if Ollama is already installed.
- It includes an embeddable chat widget, a developer API and MCP compatibility, which developers can use to build on top of it.
Cons
- Reviewers say chunking and retrieval offer little control and are hard to debug.
- Agent orchestration is shallow compared with dedicated agent frameworks.
- The desktop app is single-user, so teams need Docker or a cloud plan.
- Free users get only community support through GitHub, and an SLA starts at the $99 Pro plan.
- Cloud plans still require your own LLM API key, which is billed separately.
Key Features
AnythingLLM Pricing
Self-hosted / Desktop
- MIT open source
- Community support via GitHub
- Desktop app (single user) or Docker (multi-user)
Basic
- Custom subdomain
- Private cloud instance
- RAG and agents included
- Bring your own LLM API key
Frequently Asked Questions
Yes. The desktop app and the Docker self-hosted version are free and open source. Hosted cloud plans cost $50 or $99 a month.
No. On the Basic and Pro cloud plans you bring your own LLM API key and pay that provider separately.
SSO and RBAC are listed only on the Enterprise plan, which is priced by quote. Docker deployments have a built-in multi-user mode with role-based permissions.
Local models through Ollama, LM Studio and LocalAI, and cloud models from OpenAI, Anthropic, Azure OpenAI, Gemini, Groq, Mistral, Cohere, or any OpenAI-compatible endpoint.
Free users rely on GitHub issues and the community. The Pro plan includes a 72-hour support SLA, and Enterprise plans get a custom SLA.
Similar Tools
See all alternatives
Onyx, formerly Danswer, is an open-source AI assistant and enterprise search tool that lets staff search and chat across 50+ connected workplace apps with permissions respected.
OpenAI's top workplace tier of ChatGPT, adding agentic work, coding, company knowledge search, and enterprise security and governance controls for large organizations.

OpenAI's general-purpose AI assistant for writing, research, coding, image creation and agent tasks, sold in individual, team and enterprise plans.
Ranked near AnythingLLM in AI Software
See AnythingLLM in the listAnythingLLM is #152 in AI Software by our score.



