We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
There was an error while loading. Please reload this page.
1 parent 67c4637 commit 44607e0Copy full SHA for 44607e0
vllm/platforms/cpu.py
@@ -35,7 +35,7 @@ def get_attn_backend_cls(cls, selected_backend: _Backend, head_size: int,
35
dtype: torch.dtype, kv_cache_dtype: Optional[str],
36
block_size: int, use_v1: bool,
37
use_mla: bool) -> str:
38
- if selected_backend != _Backend.TORCH_SDPA:
+ if selected_backend and selected_backend != _Backend.TORCH_SDPA:
39
logger.info("Cannot use %s backend on CPU.", selected_backend)
40
logger.info("Using Torch SDPA backend.")
41
return "vllm.attention.backends.torch_sdpa.TorchSDPABackend"
0 commit comments