Devtoolwatch BoardAnythingLLM review
AnythingLLM is a self-hostable, local-first AI app for chatting with your documents, running AI agents, and connecting local or cloud LLMs.
TL;DR
AnythingLLM is an all-in-one AI application for chatting with documents, building agents, and running local or cloud models. It ships as a desktop app for Mac, Windows, and Linux, plus Docker and other self-hosted options. It runs locally by default, includes a built-in local LLM engine, and supports multi-user instances in the Docker version.
- 67.8kGitHub stars
- +319new stars, week of Oct 25
- #17in Inference and training this week
AnythingLLM at a glance
- Category
- Inference and training
- Maintained by
- Mintplex Labs (organization, @Mintplex-Labs on GitHub)
- License
- MIT
- Language
- JavaScript, CSS, Dockerfile
- Repository
- github.com/Mintplex-Labs/anything-llm
- Website
- anythingllm.com
- Latest release
- v1.18.0 on GitHub, Oct 13, 2026
- First published
- Jun 4, 2023
What AnythingLLM does
- AnythingLLM ingests documents such as PDF, TXT, and DOCX and lets you chat with them, with source citations.
- AnythingLLM runs local models on CPU or GPU through a built-in engine powered by Lemonade and llama.cpp.
- AnythingLLM connects to many LLM providers, including OpenAI, Anthropic, Ollama, LM Studio, AWS Bedrock, and Azure OpenAI.
- AnythingLLM includes built-in agents, a no-code agent builder, custom agents, and MCP compatibility.
- AnythingLLM runs recurring tasks or prompts on a cron schedule with full agent capabilities.
- AnythingLLM routes chats to a provider and model based on rules you define.
- AnythingLLM offers a developer API and an embeddable chat widget for websites. The widget is available in the Docker version only.
How it works
AnythingLLM runs as a desktop app or a self-hosted server, with deployment options such as Docker, AWS, GCP, and Render. The monorepo contains a Vite and React frontend, a Node.js Express server for LLM and vector database work, and a collector that parses documents. It uses LanceDB by default and also supports PGVector, Pinecone, Chroma, Qdrant, and other vector databases.
How to install and start using AnythingLLM
Requirements
- The desktop app runs on macOS, Windows, or Linux.
- GPU acceleration requires NVIDIA via CUDA, or AMD and Intel via Vulkan.
- Development setup uses yarn and requires filling in server/.env.development.
First steps
- Download the AnythingLLM desktop app for macOS, Windows, or Linux.
- Pick a model in AnythingLLM, or let it recommend one for your hardware.
- Upload documents to a workspace with drag and drop, then chat with them.
Who AnythingLLM is for
- Developers who want a private, local document-chat and agent app without cloud API keys.
- Teams that need a self-hosted, multi-user AI instance with per-user permissions.
- Builders who want to embed a chat widget or integrate through a developer API.
Limitations to know
- Multi-user support, white-labeling, and the embeddable chat widget are available in the Docker version only.
- AnythingLLM collects anonymous telemetry by default. You can turn it off with DISABLE_TELEMETRY or in the Privacy settings.
- AnythingLLM still makes outbound connections to its model CDN, GitHub, and any external providers, even with telemetry disabled.
Is AnythingLLM growing? Usage data
In the week of Oct 25–31, 2026, AnythingLLM ranked #162 on the Devtoolwatch Board and #17 in Inference and training, with a momentum score of 49 out of 100.
Show the weekly numbers
| Week of | New stars |
|---|---|
| Aug 9, 2026 | 281 |
| Aug 16, 2026 | 365 |
| Aug 23, 2026 | 321 |
| Aug 30, 2026 | 316 |
| Sep 6, 2026 | 323 |
| Sep 13, 2026 | 293 |
| Sep 20, 2026 | 313 |
| Sep 27, 2026 | 241 |
| Oct 4, 2026 | 267 |
| Oct 11, 2026 | 391 |
| Oct 18, 2026 | 293 |
| Oct 25, 2026 | 319 |
On Hacker News
- Show HN: AnythingLLM – Open-Source, All-in-One Desktop AI Assistant, 368 points and 77 comments (Sep 5, 2024)
The Board ranks tools by four signals each week: new stars, new stars for the repo's size, added downloads and download growth beyond the market. How the ranking works.
AnythingLLM alternatives
Other LLM inference and training tools on the Board in the week of Oct 25–31, 2026, by momentum:
| Rank | Tool | Stars | Downloads/wk |
|---|---|---|---|
| #1 | dsh-our-free-modeldsh-our-free-model is a dsh plugin that provides DeepSeek V4.1 Flash, Kimi K3 and other models without login, sign-up or API key. | 54.3k | – |
| #2 | magpiemagpie is a local gateway and app that lets coding agents like Claude Code, Codex and Gemini CLI use any provider's model. | 26.9k | – |
| #3 | StrataStrata is a local inference engine and app that runs large open models like Qwen3.8-Flash-Next on consumer hardware with… | 37.1k | – |
| #4 | edge0edge0 is an open-source streaming inference framework that runs large sparse mixture-of-experts models on consumer hardware by streaming… | 8.9k | – |
| #5 | OmniRouteOmniRoute is an open-source AI gateway that routes requests across many model providers through one OpenAI-compatible endpoint, with… | 83.2k | 106k |
| #6 | CLIProxyAPICLIProxyAPI is a Go proxy server that exposes CLI coding accounts like Codex, Claude Code and Gemini CLI through OpenAI, Gemini and… | 57.1k | – |
Recent AnythingLLM releases
AnythingLLM FAQ
Is AnythingLLM free and open source?
Yes. AnythingLLM's source code is on GitHub (Mintplex-Labs/anything-llm) under the MIT license, so you can use it for free.
Is AnythingLLM still maintained?
Yes. The Mintplex-Labs/anything-llm repository was last updated on Oct 25, 2026, and the latest release, v1.18.0 on GitHub, came out on Oct 13, 2026.
How popular is AnythingLLM?
AnythingLLM has 67.8k GitHub stars as of Nov 1, 2026. It gained 319 new stars in the week of Oct 25–31, 2026, against a 12-week average of 310 a week. That made it #17 among LLM inference and training tools on the Devtoolwatch Board.
AnythingLLM vs Open WebUI: how do they compare?
AnythingLLM and Open WebUI are both tracked by Devtoolwatch as LLM inference and training tools. AnythingLLM has 67.8k GitHub stars and Open WebUI has 157.4k. Open WebUI review.
AnythingLLM vs Ollama: how do they compare?
AnythingLLM and Ollama are both tracked by Devtoolwatch as LLM inference and training tools. AnythingLLM has 67.8k GitHub stars and Ollama has 185.5k. Ollama is a tool for running and building with open AI models locally or in the cloud on macOS, Windows, and Linux. Ollama review.
What are the best alternatives to AnythingLLM?
Other LLM inference and training tools on the Devtoolwatch Board in the week of Oct 25–31, 2026, by momentum: dsh-our-free-model (#1), magpie (#2), Strata (#3), edge0 (#4), OmniRoute (#5). The alternatives table above has their numbers.
Is AnythingLLM private?
AnythingLLM runs locally by default, and its homepage says document knowledge stays on your device. It collects anonymous telemetry, with no IP or other identifying data, and you can disable it. Using external LLMs, embedders, or vector databases still sends data to those providers.
How do I use AnythingLLM with Ollama or LM Studio?
AnythingLLM lists Ollama and LM Studio as supported LLM providers and as embedder options. Select one as your provider in the app to connect it.
Can AnythingLLM use my GPU?
AnythingLLM's built-in local LLM engine runs models on CPU or GPU. It supports NVIDIA through CUDA and AMD and Intel through Vulkan, with no extra setup.
Does AnythingLLM support MCP?
AnythingLLM is MCP-compatible. Its agents can use tools, and its Intelligent Skill Selection feature lets you enable many tools for a model.
Does AnythingLLM have an API?
AnythingLLM includes a full developer API for custom integrations.
How does RAG work in AnythingLLM?
AnythingLLM ingests your documents through its collector and stores embeddings in a vector database, LanceDB by default. It then answers chats with source citations. The default embedder is the AnythingLLM Native Embedder.
Sources
AnythingLLM README badge
Maintainers can show the tool's weekly Board rank in their README:
[](https://index.devtoolwatch.com/tools/anythingllm/)How this page is made: Devtoolwatch writes each review from the project's README, releases, package metadata and website, with an AI model, then checks every claim against the source it cites and drops what the sources don't support. The numbers come from Devtoolwatch's daily GitHub snapshots, GitHub's stargazer history, and the npm and PyPI download APIs. Spotted an error? Email [email protected].