Enable 4bit bnb prequant MOE (#21548)

Signed-off-by: Jee Jee Li <pandaleefree@gmail.com> Co-authored-by: Jee Jee Li <pandaleefree@gmail.com>
2025-08-11 19:02:14 -07:00
parent 1891a265d3
commit 9b94d6ec8f
2 changed files with 4 additions and 8 deletions
--- a/vllm/model_executor/models/qwen3_moe.py
+++ b/vllm/model_executor/models/qwen3_moe.py
@@ -684,4 +684,4 @@ class Qwen3MoeForCausalLM(nn.Module, SupportsPP, SupportsLoRA,
        return loader.load_weights(weights)

    def get_expert_mapping(self) -> list[tuple[str, str, int, str]]:
-        return self.model.get_expert_mapping()
+        return self.model.get_expert_mapping()