Model techniques map
Techniquesinference & servinginference scheduling

specific method · filed under inference & serving

Asynchronous scheduling

A scheduling mechanism that overlaps device-to-host sampling copies with preparation of the next decode step.

Also called async scheduling.

sources
3
models
3
labs adopt it
3
strongest
used

How sources treat it

One count per evidence span, weakest treatment to strongest.

used 3

Documented in

Further reading

Picked by hand, not extracted: where to read more, not evidence for anything on this page.

Evidence

3 spans quoted from the sources, strongest treatment first.

--async-scheduling

usedunclearin Qwen3.6-35B-A3BQwen

Asynchronous Scheduling: Within vLLM, we implemented a mechanism to overlap the 'Device-to-Host' (D2H) sampling copies with the preparation of the next decode step

usedinference servingin GLM-5Z.ai

--async-scheduling

usedsoftware implementationin Step 3.7 FlashStepFun

Filed alongside

Other methods under inference & serving :: inference scheduling.