[Model] Add LongCat-Flash (#23991)

Signed-off-by: yangxurui <yangxurui@meituan.com> Co-authored-by: yangxurui <yangxurui@meituan.com>
2025-09-25 12:53:40 +08:00
parent 90b139cfff
commit 845adb3ec6
31 changed files with 1357 additions and 66 deletions
--- a/vllm/model_executor/layers/quantization/rtn.py
+++ b/vllm/model_executor/layers/quantization/rtn.py
@@ -318,7 +318,7 @@ class RTNMoEMethod(FusedMoEMethodBase):

        from vllm.model_executor.layers.fused_moe import fused_experts

-        topk_weights, topk_ids = FusedMoE.select_experts(
+        topk_weights, topk_ids, _ = FusedMoE.select_experts(
            hidden_states=x,
            router_logits=router_logits,
            use_grouped_topk=use_grouped_topk,