[DCP] Enable TokenSpeed MLA with block-interleaved DCP (#59462)

Signed-off-by: Wei Zhao <51183510+wzhao18@users.noreply.github.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
This commit is contained in:
Wei Zhao
2026-10-03 17:26:07 +00:00
committed by GitHub
co-authored by mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
parent 4ac0d0eac2
commit 84bcbc6264
3 changed files with 156 additions and 3 deletions
+1 -1
View File
@@ -30,7 +30,7 @@ nvidia-cutlass-dsl[cu13]==4.8.0
quack-kernels==0.6.5 # Required by MSA
# Tokenspeed_MLA for faster mla with spec decode
tokenspeed-mla==0.1.8; platform_system == "Linux"
tokenspeed-mla==0.2.15; platform_system == "Linux"
# Humming kernels for quantization gemm
humming-kernels[cu13]==0.1.16