Use Open WebUI with TurboLLM
Open WebUI is a self-hosted web chat front-end. Because it speaks the OpenAI-compatible API, you can point it at TurboLLM and chat with your local model through Open WebUI's interface — no cloud, no third-party keys.
How it fits together
TurboLLM serves an OpenAI-compatible API on http://localhost:6996/v1. Open WebUI runs as a separate web app and connects to that endpoint like any other OpenAI-style provider. Once connected, TurboLLM's models show up directly in Open WebUI's model picker.
Connect Open WebUI to TurboLLM
-
Start TurboLLM with a model
Launch a model so the API is live and listening.
$turbollm launch -
Open the Connections settings in Open WebUI
In Open WebUI, go to Settings → Connections and add a new OpenAI API connection.
-
Set the base URL
Use TurboLLM's OpenAI-compatible endpoint as the base URL:
http://localhost:6996/v1 -
Add an API key (only if required)
If TurboLLM's Require API key option is on, generate a key under Developer → API Keys and paste it into the connection's API key field. If key auth is off, any placeholder value works.
-
Pick a model and chat
Save the connection. TurboLLM's models now appear in Open WebUI's model picker — select one and start chatting.
Running Open WebUI in Docker
If Open WebUI runs in Docker, localhost resolves to the Open WebUI container itself — not the host where TurboLLM is running. Point the connection at your host's LAN IP instead, and start TurboLLM bound to all interfaces so it accepts connections from the container:
turbollm launch --addr 0.0.0.0:6996Then use http://<your-host-LAN-IP>:6996/v1 as the base URL in Open WebUI.
Copy-paste snippet in the app
TurboLLM's in-app Developer screen shows a ready-made snippet for Open WebUI, so you can copy the exact base URL and key settings straight into the connection form.
See the recommended models to pick one that fits your hardware before connecting Open WebUI.