Skip to content

[FP16] Improved performance by fusing dequantize with compute in kernels: 20-30% Inference Speedup#78

Merged
mikepapadim merged 68 commits into
mainfrom
feat/deq-n-compute
Dec 11, 2025
Merged

[FP16] Improved performance by fusing dequantize with compute in kernels: 20-30% Inference Speedup#78
mikepapadim merged 68 commits into
mainfrom
feat/deq-n-compute

Remove redundant workers in Qwen3 Q8_0 FFN layers.

664f160
Select commit
Loading
Failed to load commit list.
Sign in for the full log view

The logs for this run have expired and are no longer available.