AnythingLLM
Mintplex-Labs/anything-llm
Local-first desktop and Docker AI app with document chat, agents, multi-user access and MCP support
AnythingLLM is an all-in-one AI application for chatting with your own documents and running agents on local or cloud models. Hermes Agent users can treat it as a self-hosted companion workspace with MCP compatibility, since the repository is tagged hermes-agent.
What AnythingLLM does
AnythingLLM bundles a chat interface, document ingestion, vector database support and built-in agents into one application. You connect a local or cloud language model, upload files such as PDF, TXT and DOCX, and chat with citations pointing back to the sources. It ships as a desktop app for Mac, Windows and Linux, and as a Docker version that adds multi-user support with per-user permissions and an embeddable chat widget for websites.
Beyond chat, the README lists a no-code agent builder, custom agents, scheduled tasks that run prompts on a cron schedule with agent capabilities, automatic and user-managed memories, rule-based model routing and MCP compatibility. It also describes intelligent skill selection, which it says cuts token use by up to 80 percent per query when many tools are enabled, and a Developer API for custom integrations. Supported providers include OpenAI, Anthropic, Azure OpenAI, AWS Bedrock, Google Gemini, Ollama, LM Studio, LocalAI, OpenRouter and DeepSeek.
Key features
- Chat with PDF, TXT, DOCX and other documents, with source citations
- Built-in agents that can browse the web, plus a no-code agent builder
- Scheduled tasks that run on a cron schedule with agent capabilities
- MCP compatibility and a Developer API for custom integrations
- Rule-based model routing across providers
- Multi-user permissions and an embeddable chat widget (Docker version only)
When to use it
- Running a private chat over internal documents with a local model
- Giving a team shared access to one self-hosted AI instance with per-user permissions
- Embedding a chat widget on a website using the Docker version
- Scheduling recurring prompts or agent tasks
Who it is for: People who want a self-hosted or desktop AI workspace for documents and agents, including Hermes Agent users looking for a companion app.
How it fits with Hermes Agent
The repository is tagged hermes-agent, but the opening part of the README does not describe a Hermes-specific integration. It is a general-purpose app that overlaps with Hermes Agent on MCP, memories, skills and scheduled tasks.
FAQ
What is AnythingLLM?
AnythingLLM is an open-source all-in-one AI app for chatting with your documents and running AI agents on local or cloud models. It is available as a desktop app for Mac, Windows and Linux and as a Docker version.
Does AnythingLLM work with Hermes Agent?
The README does not document a direct Hermes Agent integration, although the repository is tagged hermes-agent. It does state MCP compatibility, so the AnythingLLM MCP documentation is the place to check how it connects to external tools.
Is AnythingLLM free and open source?
Yes, the repository is released under the MIT license. The README also links a hosted instance as an alternative to running it yourself.
Similar interfaces for Hermes Agent
All interfacesDesktop app that switches API providers and manages MCP, skills and prompts across coding agents
nexu-io OpenDesignLocal-first design workspace that turns your coding agent, Hermes Agent included, into a design tool
stablyai OrcaDesktop and mobile workspace for running parallel coding agents, including Hermes Agent, in git worktrees
CherryHQ Cherry StudioCross-platform desktop client for many LLM providers with 300+ assistants and MCP support
mindsdb MindsHubAgent workspace with swappable open-source agent harnesses, including Hermes, and a model router
iOfficeAI AionUiOpen-source desktop and WebUI Cowork app that runs Hermes, Claude Code, Codex and 20+ other CLI agents
Related guides: Connect Hermes agents on several machines with Hermes Desktop · How to install Hermes Agent