fix amd triton mla path (#17871)

2025-05-11 00:53:31 -07:00
parent 9112155283
commit eea22a56ab
1 changed files with 1 additions and 1 deletions
--- a/vllm/attention/backends/mla/common.py
+++ b/vllm/attention/backends/mla/common.py
@ -1063,7 +1063,7 @@ class MLACommonImpl(MLAAttentionImpl[T], Generic[T]):
                softmax_scale,
                None,  # bias
            )
-        if is_vllm_fa:
+        elif is_vllm_fa:
            attn_out = self.flash_attn_varlen_func(
                q=q,
                k=k,