vllm_omni.worker.gpu_ar_worker ¶
GPUARWorker ¶
Bases: OmniWorkerMixin, OmniGPUWorkerBase
GPU worker for autoregressive omni model stages.
Extends the base GPUWorker to initialize and manage autoregressive model runners for text generation stages (e.g., thinker stages).
handle_sleep_task ¶
handle_sleep_task(task: OmniSleepTask | dict) -> OmniACK
Explicitly handle sleep commands. Calls the implementation in the base class OmniGPUWorkerBase.
handle_wake_task ¶
handle_wake_task(task: OmniWakeTask | dict) -> OmniACK
Explicitly handle wake-up commands. Calls the implementation in the base class OmniGPUWorkerBase.