Skip to content

vllm_omni.worker_v2.native_output_worker ¶

Materialize native producer output independently of scheduler consumption.

NativeAsyncOutput ¶

Bases: AsyncModelRunnerOutput

Consume an ordered publication on the engine's owner thread.

copy_event `instance-attribute` ¶

copy_event = copy_event

get_output ¶

get_output() -> Any

NativeOutputWorker ¶

Bounded FIFO CPU materialization and native-queue publication.

Input outputs must have no owner-thread finalizer installed. Reservations and metadata writes must be complete before submit. This worker never updates scheduler state or drains connector notifications.

close ¶

close() -> None

submit ¶

submit(
    output: AsyncModelRunnerOutput, plane: Any
) -> NativeAsyncOutput