biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[BugFix] Fix architecture flags to prevent issues on SM103 (#31150)

Browse Source 
Signed-off-by: LopezCastroRoberto <robertol.c510@gmail.com>
This commit is contained in:
Roberto L. Castro
2026-01-05 21:11:35 +01:00
committed by GitHub
parent 5708297e4e
commit fdcc5176be
2 changed files with 13 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
tools/flashinfer-build.sh
View File

@@ -32,9 +32,12 @@ if [[ "${CUDA_VERSION}" == 11.* ]]; then
FI_TORCH_CUDA_ARCH_LIST="7.5 8.0 8.9"
elif [[ "${CUDA_VERSION}" == 12.[0-7]* ]]; then
FI_TORCH_CUDA_ARCH_LIST="7.5 8.0 8.9 9.0a"
elif [[ "${CUDA_VERSION}" == 12.[8-9]* ]]; then
# CUDA 12.8–12.9
FI_TORCH_CUDA_ARCH_LIST="7.5 8.0 8.9 9.0a 10.0a 10.3a 12.0"
else
# CUDA 12.8+ supports 10.0a and 12.0
FI_TORCH_CUDA_ARCH_LIST="7.5 8.0 8.9 9.0a 10.0a 12.0"
# CUDA 13.0+
FI_TORCH_CUDA_ARCH_LIST="7.5 8.0 8.9 9.0a 10.0f 12.0"
fi
echo "🏗️ Building FlashInfer AOT for arches: ${FI_TORCH_CUDA_ARCH_LIST}"