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

Fix INT8 quantization error on Blackwell GPUs (SM100+) (#25935)

Browse Source 
Signed-off-by: padg9912 <phone.and.desktop@gmail.com>
This commit is contained in:
Param
2025-09-30 22:19:53 -04:00
committed by GitHub
parent 1244948885
commit 99028fda44
2 changed files with 9 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
csrc/quantization/cutlass_w8a8/c3x/scaled_mm_helper.hpp
View File

@ -25,7 +25,10 @@ void dispatch_scaled_mm(torch::Tensor& c, torch::Tensor const& a,
if constexpr (!std::is_same_v<Int8Func, std::nullptr_t>) {
int8_func(c, a, b, a_scales, b_scales, bias);
} else {
TORCH_CHECK(false, "Int8 not supported for this architecture");
int32_t version_num = get_sm_version_num();
TORCH_CHECK(
false, "Int8 not supported on SM", version_num,
". Use FP8 quantization instead, or run on older arch (SM < 100).");
}
}
} else {