[moe][quant] add weight name case for offset (#15515)

Signed-off-by: Mengqing Cao <cmq0113@163.com>
2025-03-27 12:50:29 +08:00
parent 7f301dd8ef
commit fb22be5817
1 changed files with 3 additions and 2 deletions
--- a/vllm/model_executor/layers/fused_moe/layer.py
+++ b/vllm/model_executor/layers/fused_moe/layer.py
@ -699,8 +699,9 @@ class FusedMoE(torch.nn.Module):
                             tp_rank=self.tp_rank)
            return

-        # Case weight scales and zero_points
-        if ("scale" in weight_name or "zero" in weight_name):
+        # Case weight scales, zero_points and offset
+        if ("scale" in weight_name or "zero" in weight_name
+                or "offset" in weight_name):
            # load the weight scales and zp based on the quantization scheme
            # supported weight scales/zp can be found in
            # FusedMoeWeightScaleSupported