Skip to content

[CPU] Optimize GQA attention bias application for FP16#25871

Merged
derdeljan-msft merged 1 commit intomainfrom
derdeljan/optimize_gqa_spec_decoding_fp16
Aug 28, 2025
Merged

[CPU] Optimize GQA attention bias application for FP16#25871
derdeljan-msft merged 1 commit intomainfrom
derdeljan/optimize_gqa_spec_decoding_fp16

Commits

Commits on Aug 27, 2025