AnythingLLM Guide

AnythingLLM can expose an OpenAI-compatible API for chatting with your document workspace from 3sparks Chat.

Connect AnythingLLM

1. Start AnythingLLM and confirm its API endpoint is enabled.
2. On iPhone or iPad, use a LAN IP, remote host, or Tailscale IP that can reach the AnythingLLM server.
3. In 3sparks Chat, open Connections and tap +.
4. Choose the AnythingLLM provider or Other (Custom) if needed.
5. Leave the name blank for a generated name, or enter your own.
6. Enter the base URL and route prefix shown by AnythingLLM.
7. Add an API token if your AnythingLLM server requires one.
8. Tap Create Connection, then fetch or enter a model.

Troubleshooting

• Do not use localhost from a phone unless the server runs on that phone.
• Confirm the route prefix matches the OpenAI-compatible endpoint.
• If authentication is enabled, add the token to the connection.
• If model fetch fails, check AnythingLLM workspace/API settings.