fix: GLM4.7-FP4 usage

python/sglang/srt/layers/quantization/base_config.py

@@ -173,12 +173,10 @@ def _modelopt_override_quantization_method(
         # Check if this is a ModelOpt config
         quant_algo = hf_quant_config.get("quant_algo", "").upper()
 
-        # If user specified generic "modelopt", auto-detect the specific method
-        if user_quant == "modelopt":
-            if "FP8" in quant_algo:
-                return "modelopt_fp8"
-            elif "NVFP4" in quant_algo or "FP4" in quant_algo:
-                return "modelopt_fp4"
+        if "FP8" in quant_algo:
+            return "modelopt_fp8"
+        elif "NVFP4" in quant_algo or "FP4" in quant_algo:
+            return "modelopt_fp4"
 
         # The hf_quant_config may be a parsed quant config, so we need to check the
         # quant_method.

python/sglang/srt/layers/quantization/modelopt_quant.py

@@ -1513,7 +1513,7 @@ def _slice_scale(w):
                 # For other backends, ensure the per-input block dimension is aligned to 16.
                 assert (
                     weight_scale.shape[assert_dim] % block_size == 0
-                ), f"Expected {name}_weight_scale.dim({assert_dim}) to be divisible by {block_size}"
+                ), f"Expected {name}_weight_scale.dim({assert_dim}) to be divisible by {block_size}, got {weight_scale.shape[assert_dim]}"
             assert (
                 weight_scale.dtype == torch.float8_e4m3fn
             ), f"{name} Weight Blockscale must be represented as FP8-E4M3"

python/sglang/srt/models/glm4_moe.py

@@ -63,6 +63,7 @@
 from sglang.srt.layers.moe.ep_moe.layer import get_moe_impl_class
 from sglang.srt.layers.moe.fused_moe_triton.layer import FusedMoE
 from sglang.srt.layers.moe.topk import TopK
+from sglang.srt.layers.moe.utils import RoutingMethodType
 from sglang.srt.layers.quantization.base_config import QuantizationConfig
 from sglang.srt.layers.quantization.fp8_kernel import is_fp8_fnuz
 from sglang.srt.layers.radix_attention import RadixAttention
@@ -375,6 +376,7 @@ def __init__(
             intermediate_size=config.moe_intermediate_size,
             quant_config=quant_config,
             routed_scaling_factor=self.routed_scaling_factor,
+            routing_method_type=RoutingMethodType.DeepSeekV3,
             prefix=add_prefix("experts", prefix),
         )
 

python/sglang/srt/models/glm4_moe_nextn.py

@@ -21,7 +21,8 @@
 from torch import nn
 from transformers import PretrainedConfig
 
-from sglang.srt.distributed import get_tensor_model_parallel_world_size
+from sglang.srt.distributed import get_pp_group, get_tensor_model_parallel_world_size
+from sglang.srt.environ import envs
 from sglang.srt.eplb.expert_distribution import get_global_expert_distribution_recorder
 from sglang.srt.layers.dp_attention import is_dp_attention_enabled
 from sglang.srt.layers.layernorm import RMSNorm
@@ -34,11 +35,14 @@
 from sglang.srt.model_executor.forward_batch_info import ForwardBatch
 from sglang.srt.models.glm4_moe import Glm4MoeDecoderLayer, Glm4MoeForCausalLM
 from sglang.srt.server_args import get_global_server_args
-from sglang.srt.utils import add_prefix
+from sglang.srt.utils import add_prefix, is_cuda
 
 logger = logging.getLogger(__name__)
 
 
+_is_cuda = is_cuda()
+
+
 class Glm4MoeModelNextN(nn.Module):
     def __init__(
         self,
@@ -67,12 +71,19 @@ def __init__(
 
         self.eh_proj = nn.Linear(2 * config.hidden_size, config.hidden_size, bias=False)
 
+        self.alt_stream = (
+            torch.cuda.Stream()
+            if _is_cuda or envs.SGLANG_NPU_USE_MULTI_STREAM.get()
+            else None
+        )
+
         self.decoder = Glm4MoeDecoderLayer(
             config,
             0,
             quant_config=quant_config,
             is_nextn=True,
             prefix=add_prefix("decoder", prefix),
+            alt_stream=self.alt_stream,
         )
 
         self.shared_head = nn.Module()
@@ -127,6 +138,9 @@ def __init__(
         self.config = config
         self.tp_size = get_tensor_model_parallel_world_size()
         self.quant_config = quant_config
+        self.pp_group = get_pp_group()
+        self.num_fused_shared_experts = 0
+        self.determine_num_fused_shared_experts()
         self.model = Glm4MoeModelNextN(
             config, quant_config, prefix=add_prefix("model", prefix)
         )
@@ -139,10 +153,6 @@ def __init__(
         )
         self.logits_processor = LogitsProcessor(config)
 
-        self.num_fused_shared_experts = (
-            0 if get_global_server_args().disable_shared_experts_fusion else 1
-        )
-
     @torch.no_grad()
     def forward(
         self,