A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 14, 04:15 UTC
[Build] Revert triton_kernels requirements (#27659)
Varun Sundar Rabindranath committed
Oct 29, 2025 at 00:07 UTC
f6d5f5888cce05b4d069e2d26503934ddf6e25d4
1 file changed
-2
requirements/cuda.txt
-2
@@ -13,5 +13,3 @@ torchvision==0.24.0 # Required for phi3v processor. See https://github.com/pytor
13
# xformers==0.0.32.post1; platform_system == 'Linux' and platform_machine == 'x86_64' # Requires PyTorch >= 2.8
14
# FlashInfer should be updated together with the Dockerfile
15
flashinfer-python==0.4.1
16
-# Triton Kernels are needed for mxfp4 fused moe. (Should be updated alongside torch)
17
-triton_kernels @ git+https://github.com/triton-lang/triton.git@v3.5.0#subdirectory=python/triton_kernels