Repository navigation

Add a prebuilt DeepGEMM option to vLLM fake-quant serving - #2717

Closed

sychen52 wants to merge 2 commits into

NVIDIA:mainfrom

sychen52:vllm-prebuilt-deepgemm

Closed

Add a prebuilt DeepGEMM option to vLLM fake-quant serving#2717
sychen52 wants to merge 2 commits into
NVIDIA:mainfrom
sychen52:vllm-prebuilt-deepgemm

Commits on Oct 8, 2026
- Add a prebuilt DeepGEMM option to vLLM fake-quant serving
  sychen52
  committed
  b656d05
- Install a DeepGEMM source tree with pip instead of DEEPGEMM_URL
  sychen52
  committed
  6d317f3