Skip to content
CI

【Inference Optimize】Optimize the repeated calculation part in the DSK attention layer and optimize the performance under mixed batch. #2061

Sign in to view logs

【Inference Optimize】Optimize the repeated calculation part in the DSK attention layer and optimize the performance under mixed batch.

【Inference Optimize】Optimize the repeated calculation part in the DSK attention layer and optimize the performance under mixed batch. #2061

build

build
succeeded Aug 11, 2025 in 3s