Add DCP + Eagle support for Tokenspeed MLA backends (#48180)

Signed-off-by: Pavani Majety <pmajety@nvidia.com>
Signed-off-by: Jingyi Yang <girasoleyang@gmail.com>
Co-authored-by: Jingyi Yang <girasoleyang@gmail.com>
This commit is contained in:
Pavani Majety
2026-07-13 11:46:02 -07:00
committed by GitHub
co-authored by Jingyi Yang
parent 18c4067a54
commit 7fc97042c3
8 changed files with 313 additions and 23 deletions
+1 -1
View File
@@ -26,7 +26,7 @@ nvidia-cutlass-dsl[cu13]==4.5.2
quack-kernels>=0.3.3
# Tokenspeed_MLA for faster mla with spec decode
tokenspeed-mla==0.1.2; platform_system == "Linux"
tokenspeed-mla==0.1.8; platform_system == "Linux"
# Humming kernels for quantization gemm
humming-kernels[cu13]==0.1.10