AnythingLLM
AnythingLLM is an open-source, all-in-one AI application from Mintplex Labs that lets users chat with their own documents, run multi-agent workflows, and connect a choice of LLM and vector database backends from one self-hostable interface. It ships as both a desktop app and a Docker-deployable server.
What gets deployed with it
- Qdrant — required First dependency target for vector retrieval and document embedding storage.
Baseline topologies
| Topology | Minimum nodes | Roles |
|---|---|---|
| local-all-in-one | 1 | anythingllm |
| shared-rag-service | 2 | anythingllm, anythingllm |
What drives the cost
- Active knowledge workspaces
- Approximate number of active knowledge workspaces or teams served by the deployment.
- Indexed documents (millions)
- Approximate number of indexed documents across all workspaces.
- Peak concurrent chats
- Expected peak number of simultaneous chat or retrieval sessions.
Where it runs
bare_metal, dedicated, cloud_vm
Size AnythingLLM and get a cost estimate