Open Models Almanac
Methodspost-trainingrollout & RL infrastructure

single mention · post-training

Non-blocking weight synchronization to inference

Synchronizes updated weights to the inference engine without waiting for ongoing generations to finish.

The sources on this page only mention it in passing. The line above is the extraction's one-line gloss. See rollout & RL infrastructure for the methods filed beside it.

Filed under post-training › rollout & RL infrastructure