[NPU] bugfix for Qwen3-Next and performance update (#11969)

.github/workflows/release-docker-npu-nightly.yml

@@ -73,6 +73,6 @@ jobs:
           push: ${{ github.repository == 'sgl-project/sglang' && github.event_name != 'pull_request' }}
           provenance: false
           build-args: |
-            SGLANG_KERNEL_NPU_TAG=20250926
+            SGLANG_KERNEL_NPU_TAG=20251030
             CANN_VERSION=${{ matrix.cann_version }}
             DEVICE_TYPE=${{ matrix.device_type }}

.github/workflows/release-docker-npu.yml

@@ -69,6 +69,6 @@ jobs:
           push: ${{ github.repository == 'sgl-project/sglang' && github.event_name != 'pull_request' }}
           provenance: false
           build-args: |
-            SGLANG_KERNEL_NPU_TAG=20250926
+            SGLANG_KERNEL_NPU_TAG=20251030
             CANN_VERSION=${{ matrix.cann_version }}
             DEVICE_TYPE=${{ matrix.device_type }}

python/sglang/srt/layers/attention/fla/layernorm_gated.py

@@ -12,7 +12,9 @@ import triton
 import triton.language as tl
 from einops import rearrange
 
-from sglang.srt.utils import device_context
+from sglang.srt.utils import device_context, is_npu
+
+_is_npu = is_npu()
 
 
 def rms_norm_ref(
@@ -182,6 +184,10 @@ def _layer_norm_fwd(
     return out, mean, rstd
 
 
+if _is_npu:
+    from sgl_kernel_npu.fla.layernorm_gated import layer_norm_fwd_npu as _layer_norm_fwd
+
+
 def rms_norm_gated(
     *,
     x,

python/sglang/srt/layers/attention/mamba/mamba.py

@@ -13,16 +13,6 @@ from sglang.srt.distributed import (
     get_tensor_model_parallel_world_size,
 )
 from sglang.srt.distributed.utils import divide
-from sglang.srt.layers.attention.mamba.causal_conv1d import (
-    causal_conv1d_fn,
-    causal_conv1d_update,
-)
-from sglang.srt.layers.attention.mamba.causal_conv1d_triton import (
-    causal_conv1d_fn as causal_conv1d_fn_triton,
-)
-from sglang.srt.layers.attention.mamba.causal_conv1d_triton import (
-    causal_conv1d_update as causal_conv1d_update_triton,
-)
 from sglang.srt.layers.attention.mamba.mamba2_metadata import Mamba2Metadata
 from sglang.srt.layers.attention.mamba.mixer2_rms_norm_gated import Mixer2RMSNormGated
 from sglang.srt.layers.attention.mamba.ops import (
@@ -40,7 +30,26 @@ from sglang.srt.model_loader.weight_utils import (
     composed_weight_loader,
     sharded_weight_loader,
 )
-from sglang.srt.utils import set_weight_attrs
+from sglang.srt.utils import is_cuda, is_npu, set_weight_attrs
+
+if is_cuda():
+    from sglang.srt.layers.attention.mamba.causal_conv1d import (
+        causal_conv1d_fn,
+        causal_conv1d_update,
+    )
+    from sglang.srt.layers.attention.mamba.causal_conv1d_triton import (
+        causal_conv1d_fn as causal_conv1d_fn_triton,
+    )
+    from sglang.srt.layers.attention.mamba.causal_conv1d_triton import (
+        causal_conv1d_update as causal_conv1d_update_triton,
+    )
+elif is_npu():
+    from sgl_kernel_npu.mamba.causal_conv1d import (
+        causal_conv1d_fn_npu as causal_conv1d_fn,
+    )
+    from sgl_kernel_npu.mamba.causal_conv1d import (
+        causal_conv1d_update_npu as causal_conv1d_update,
+    )
 
 LoaderFunction = Callable[[torch.Tensor, torch.Tensor], None]
 

python/sglang/srt/layers/moe/topk.py

@@ -314,16 +314,41 @@ class TopK(CustomOp):
         num_token_non_padded: Optional[torch.Tensor] = None,
         expert_location_dispatch_info: Optional[ExpertLocationDispatchInfo] = None,
     ) -> TopKOutput:
-        global_num_experts = router_logits.shape[-1]
 
-        # NOTE: now npu_moe_gating_top_k can only support `group_count=256` pattern
-        if global_num_experts == 256:
+        use_grouped_topk = self.topk_config.use_grouped_topk
+        torch_native = self.topk_config.torch_native
+        renormalize = self.topk_config.renormalize
+
+        if not use_grouped_topk and not torch_native:
+            topk_weights, topk_ids, _ = torch_npu.npu_moe_gating_top_k_softmax(
+                router_logits,
+                k=self.topk_config.top_k,
+            )
+            topk_weights = topk_weights.to(torch.float32)
+
+            if renormalize:
+                topk_weights_sum = (
+                    topk_weights.sum(dim=-1, keepdim=True)
+                    if self.topk_config.num_fused_shared_experts == 0
+                    else topk_weights[:, :-1].sum(dim=-1, keepdim=True)
+                )
+                topk_weights = topk_weights / topk_weights_sum
 
+            if expert_location_dispatch_info is not None:
+                topk_ids = topk_ids_logical_to_physical(
+                    topk_ids, expert_location_dispatch_info
+                )
+            get_global_expert_distribution_recorder().on_select_experts(
+                topk_ids=topk_ids
+            )
+
+            return StandardTopKOutput(topk_weights, topk_ids, _)
+        if use_grouped_topk and not torch_native and router_logits.shape[-1] == 256:
+            # NOTE: now npu_moe_gating_top_k can only support `group_count=256` pattern
             routed_scaling_factor = self.topk_config.routed_scaling_factor or 1
-            router_logits = router_logits.to(torch.float32)
 
             topk_weights, topk_ids, _ = torch_npu.npu_moe_gating_top_k(
-                router_logits,
+                router_logits.to(torch.float32),
                 k=self.topk_config.top_k,
                 bias=self.topk_config.correction_bias.to(torch.float32),
                 k_group=self.topk_config.topk_group,
@@ -335,7 +360,7 @@ class TopK(CustomOp):
                 eps=float(1e-20),
             )
 
-            if self.topk_config.renormalize:
+            if renormalize:
                 topk_weights_sum = (
                     topk_weights.sum(dim=-1, keepdim=True)
                     if self.topk_config.num_fused_shared_experts == 0

python/sglang/srt/models/qwen3_next.py

@@ -478,6 +478,13 @@ class Qwen3GatedDeltaNet(nn.Module):
         # reshape input data into 2D tensor
         core_attn_out = core_attn_out.reshape(-1, core_attn_out.shape[-1])
         z = z.reshape(-1, z.shape[-1])
+
+        # Add padding for DP-Attn
+        if is_dp_attention_enabled():
+            core_attn_out_pad = torch.zeros_like(z)
+            core_attn_out_pad[: core_attn_out.shape[0], :] = core_attn_out
+            core_attn_out = core_attn_out_pad
+
         core_attn_out = self.norm(core_attn_out, z)
         core_attn_out = core_attn_out.reshape(z_shape_og)
         core_attn_out = core_attn_out.reshape(*core_attn_out.shape[:-2], -1)

scripts/ci/npu_ci_install_dependency.sh

@@ -59,7 +59,7 @@ wget -O "${BISHENG_NAME}" "${BISHENG_URL}" && chmod a+x "${BISHENG_NAME}" && "./
 
 
 ### Install sgl-kernel-npu
-SGL_KERNEL_NPU_TAG="20250926"
+SGL_KERNEL_NPU_TAG="20251030"
 git clone --depth 1 https://github.com/sgl-project/sgl-kernel-npu.git --branch ${SGL_KERNEL_NPU_TAG}
 # pin wheel to 0.45.1 ref: https://github.com/pypa/wheel/issues/662
 pip install wheel==0.45.1