Update to new version of base image

2025-10-24 15:45:06 +08:00
parent ee04aead1e
commit fad74b701b
476 changed files with 1270 additions and 46 deletions
--- a/vllm/model_executor/layers/linear.py
+++ b/vllm/model_executor/layers/linear.py
@@ -132,7 +132,7 @@ class UnquantizedLinearMethod(LinearMethodBase):
              layer: torch.nn.Module,
              x: torch.Tensor,
              bias: Optional[torch.Tensor] = None) -> torch.Tensor:
-        if (x.shape[0] == 16384 or x.shape[0] == 15360):
+        if (x.shape[0] == 8192 or x.shape[0] == 16384 or x.shape[0] == 15360):
            if bias is None:
                return x @ layer.weight.T
            else: