NexoRouter

Documentation

LiteLLM in the NexoRouter documentation.

LiteLLM

Status: Candidate setup, not yet verified by NexoRouter.

LiteLLM can route OpenAI-compatible requests to custom endpoints. Use it carefully because it can hide which endpoint, model, or key is actually used.

Configuration values

Base URL: https://api.nexorouter.com/v1
API key: YOUR_NEXOROUTER_API_KEY
Model: deepseek-v4-flash

Verification path

  1. Configure one model only.
  2. Send one non-streaming chat request.
  3. Check LiteLLM logs and NexoRouter Usage Logs.
  4. Add fallback or routing rules only after the single-model path is proven.

Risks

  • Double routing can make debugging harder.
  • Fallback models can mask model_not_found.
  • Streaming and tool calling need separate checks.
LiteLLM — NexoRouter