Deploy AnythingLLM on Your VPS
Deploy AnythingLLM on your VPS in one click. An all-in-one RAG workspace — drop in documents and ask questions grounded in them, with its own vector database built in.
AnythingLLM turns your documents into something you can ask questions of. Upload PDFs, Word files, spreadsheets, or scrape a website into a workspace, and chat against that content with citations back to the source. Workspaces keep contexts separate — one per client, project, or team — and you can invite colleagues with their own accounts and permissions.
Vessl deploys it as a single service: it embeds LanceDB as its vector store and SQLite for metadata, both on one persistent volume, so there is no separate database to run. Point it at the Ollama template for a fully local, fully private setup, or plug in OpenAI, Anthropic, or any OpenAI-compatible endpoint. The first account you create becomes the admin.
What's included
- Single service — vector database and metadata store are built in
- Upload PDFs, Word, spreadsheets, or scrape a site into a workspace
- Answers cite the source documents they came from
- Separate workspaces with per-user permissions for a team
- Works with Ollama for fully local AI, or OpenAI / Anthropic
Common use cases
- Ask questions across your own documentation, contracts, or research
- Private "chat with our handbook" for a team, on your own server
- A RAG prototype without assembling a vector DB and pipeline yourself
Frequently asked questions
Do I need a separate vector database?
No. AnythingLLM embeds LanceDB and stores everything on its volume, which is why this template is one service. You can point it at an external vector database later if you outgrow that.
Can I keep everything local?
Yes — deploy the Ollama template alongside it and select Ollama as the LLM provider. Nothing then leaves your server. Telemetry is disabled by default in this template.
How is this different from Open WebUI?
Open WebUI is primarily a chat interface for models. AnythingLLM is built around documents: workspaces, ingestion, and citation-backed answers over your own files. Plenty of people run both.
Ready to ship?
Deploy AnythingLLM in under a minute.
Connect your VPS, pick this template, fill in any required fields. Vessl handles the container, SSL, and persistent storage.
Start for FreeNo credit card · BYOS · IDR billing