fix SUPPORT_CUTLASS_BLOCK_FP8 flag (#4640)

2025-03-21 00:45:07 -04:00
parent ad4e58bf67
commit 7b5fc71972
1 changed files with 1 additions and 1 deletions
--- a/python/sglang/srt/layers/quantization/fp8_utils.py
+++ b/python/sglang/srt/layers/quantization/fp8_utils.py
@@ -82,7 +82,7 @@ def normalize_e4m3fn_to_e4m3fnuz(


 def cutlass_block_fp8_supported() -> bool:
-    if get_bool_env_var("SUPPORT_CUTLASS_BLOCK_FP8"):
+    if not get_bool_env_var("SUPPORT_CUTLASS_BLOCK_FP8"):
        return False
    if _is_cuda:
        major, minor = torch.cuda.get_device_capability()