ray.serve.llm.LLMServer.resume

async LLMServer.resume(**kwargs: Any) → None

Resume generation on the engine after pause.

Parameters:

**kwargs – Engine-specific resume options. Passed through to the engine.