AnythingLLM: A private ChatGPT to chat with anything!.
An efficient, customizable, and open-source enterprise-ready document chatbot solution.
| | Docs | Hosted Instance
A full-stack application that enables you to turn any document, resource, or piece of content into context that any LLM can use as references during chatting. This application allows you to pick and choose which LLM or Vector Database you want to use as well as supporting multi-user management and permissions. ![Chatting](/images/screenshots/chatting.gif) ### Watch the demo! [![Watch the video](/images/youtube.png)](https://youtu.be/f95rGD9trL0) ### Product Overview AnythingLLM is a full-stack application where you can use commercial off-the-shelf LLMs or popular open source LLMs and vectorDB solutions to build a private ChatGPT with no compromises that you can run locally as well as host remotely and be able to chat intelligently with any documents you provide it. AnythingLLM divides your documents into objects called `workspaces`. A Workspace functions a lot like a thread, but with the addition of containerization of your documents. Workspaces can share documents, but they do not talk to each other so you can keep your context for each workspace clean. Some cool features of AnythingLLM - **Multi-user instance support and permissioning** - Multiple document type support (PDF, TXT, DOCX, etc) - Manage documents in your vector database from a simple UI - Two chat modes `conversation` and `query`. Conversation retains previous questions and amendments. Query is simple QA against your documents - In-chat citations linked to the original document source and text - Simple technology stack for fast iteration - 100% Cloud deployment ready. - "Bring your own LLM" model. - Extremely efficient cost-saving measures for managing very large documents. You'll never pay to embed a massive document or transcript more than once. 90% more cost effective than other document chatbot solutions. - Full Developer API for custom integrations! ### Supported LLMs, Embedders, and Vector Databases **Supported LLMs:** - [Any open-source llama.cpp compatible model](/server/storage/models/README.md#text-generation-llm-selection) - [OpenAI](https://openai.com) - [Azure OpenAI](https://azure.microsoft.com/en-us/products/ai-services/openai-service) - [Anthropic ClaudeV2](https://www.anthropic.com/) - [LM Studio (all models)](https://lmstudio.ai) - [LocalAi (all models)](https://localai.io/) **Supported Embedding models:** - [AnythingLLM Native Embedder](/server/storage/models/README.md) (default) - [OpenAI](https://openai.com) - [Azure OpenAI](https://azure.microsoft.com/en-us/products/ai-services/openai-service) - [LM Studio (all)](https://lmstudio.ai) - [LocalAi (all)](https://localai.io/) **Supported Vector Databases:** - [LanceDB](https://github.com/lancedb/lancedb) (default) - [Pinecone](https://pinecone.io) - [Chroma](https://trychroma.com) - [Weaviate](https://weaviate.io) - [QDrant](https://qdrant.tech) ### Technical Overview This monorepo consists of three main sections: - `collector`: Python tools that enable you to quickly convert online resources or local documents into LLM useable format. - `frontend`: A viteJS + React frontend that you can run to easily create and manage all your content the LLM can use. - `server`: A nodeJS + express server to handle all the interactions and do all the vectorDB management and LLM interactions. - `docker`: Docker instructions and build process + information for building from source. ### Minimum Requirements > [!TIP] > Running AnythingLLM on AWS/GCP/Azure? > You should aim for at least 2GB of RAM. Disk storage is proprotional to however much data > you will be storing (documents, vectors, models, etc). Minimum 10GB recommended. - `yarn` and `node` on your machine - `python` 3.9+ for running scripts in `collector/`. - access to an LLM running locally or remotely. *AnythingLLM by default uses a built-in vector database powered by [LanceDB](https://github.com/lancedb/lancedb) *AnythingLLM by default embeds text on instance privately [Learn More](/server/storage/models/README.md) ## Recommended usage with Docker (easy!) > [!TIP] > It is best to mount the containers storage volume to a folder on your host machine > so that you can pull in future updates without deleting your existing data! `docker pull mintplexlabs/anythingllm:master` ```shell export STORAGE_LOCATION="/var/lib/anythingllm" && \ mkdir -p $STORAGE_LOCATION && \ touch "$STORAGE_LOCATION/.env" && \ docker run -d -p 3001:3001 \ -v ${STORAGE_LOCATION}:/app/server/storage \ -v ${STORAGE_LOCATION}/.env:/app/server/.env \ -e STORAGE_DIR="/app/server/storage" \ mintplexlabs/anythingllm:master ``` Open [http://localhost:3001](http://localhost:3001) and you are now using AnythingLLM! All your data and progress will now persist between container rebuilds or pulls from Docker Hub. [Learn more about running AnythingLLM with Docker](./docker/HOW_TO_USE_DOCKER.md) ### How to get started (Development environment) - `yarn setup` from the project root directory. - This will fill in the required `.env` files you'll need in each of the application sections. Go fill those out before proceeding or else things won't work right. - `yarn prisma:setup` To build the Prisma client and migrate the database. To boot the server locally (run commands from root of repo): - ensure `server/.env.development` is set and filled out. `yarn dev:server` To boot the frontend locally (run commands from root of repo): - ensure `frontend/.env` is set and filled out. - ensure `VITE_API_BASE="http://localhost:3001/api"` `yarn dev:frontend` [Learn about documents](./server/storage/documents/DOCUMENTS.md) [Learn about vector caching](./server/storage/vector-cache/VECTOR_CACHE.md) ## Standalone scripts This repo contains standalone scripts you can run to collect data from a Youtube Channel, Medium articles, local text files, word documents, and the list goes on. This is where you will use the `collector/` part of the repo. [Go set up and run collector scripts](./collector/README.md) ## Contributing - create issue - create PR with branch name format of `