LiteLLM

LiteLLM lets RealTimeX use an OpenAI-compatible gateway that can fan out to many upstream providers behind one endpoint.

Open it from Settings > AI Providers > LLM, then choose LiteLLM.

What you configure

The current LiteLLM page exposes:

• Base URL
• Chat model selection
• Token context window
• an optional API key

Current setup flow

1. Open Settings > AI Providers > LLM.
2. Choose LiteLLM.
3. Enter the LiteLLM base URL.
4. Add the API key if your LiteLLM gateway requires one.
5. Wait for the model list to load.
6. Select the routed model you want RealTimeX to use.
7. Set the token context window you want RealTimeX to assume.
8. Save the provider settings.
9. Test a normal chat.

When to use LiteLLM

• Use it when you want one gateway in front of multiple vendors.
• Use it when you need centralized routing, budgets, or auth handling outside RealTimeX.
• Use it when a team standardizes on OpenAI-compatible proxy layers.