import enum import logging from typing import Any, Iterable, Optional, Set, Tuple import torch import triton from torch import nn from sglang.jit_kernel.triton.gdn_fused_proj import fused_qkvzba_split_reshape_cat from sglang.srt.configs.qwen3_next import Qwen3NextConfig from sglang.srt.distributed import get_pp_group from sglang.srt.eplb.expert_distribution import get_global_expert_distribution_recorder from sglang.srt.eplb.expert_location import ModelConfigForExpertLocation from sglang.srt.layers.attention.fla.fused_norm_gate import FusedRMSNormGated from sglang.srt.layers.attention.fla.layernorm_gated import RMSNorm as RMSNormGated from sglang.srt.layers.attention.mamba.mamba import mamba_v2_sharded_weight_loader from sglang.srt.layers.communicator import LayerCommunicator, LayerScatterModes from sglang.srt.layers.dp_attention import ( is_dp_attention_enabled, ) from sglang.srt.layers.layernorm import GemmaRMSNorm from sglang.srt.layers.linear import ( ColumnParallelLinear, MergedColumnParallelLinear, QKVParallelLinear, RowParallelLinear, ) from sglang.srt.layers.logits_processor import LogitsProcessor from sglang.srt.layers.moe.fused_moe_triton.layer import FusedMoE from sglang.srt.layers.quantization.base_config import QuantizationConfig from sglang.srt.layers.radix_attention import RadixAttention from sglang.srt.layers.radix_linear_attention import RadixLinearAttention from sglang.srt.layers.rotary_embedding import get_rope from sglang.srt.layers.vocab_parallel_embedding import ( ParallelLMHead, VocabParallelEmbedding, ) from sglang.srt.model_executor.cuda_graph_config import ( Backend, Phase, check_cuda_graph_backend, ) from sglang.srt.model_executor.forward_batch_info import ForwardBatch from sglang.srt.model_executor.runner import get_is_capture_mode from sglang.srt.model_loader.weight_utils import ( default_weight_loader, sharded_weight_loader, ) from sglang.srt.models.qwen2_moe import Qwen2MoeMLP, Qwen2MoeSparseMoeBlock from sglang.srt.runtime_context import ( get_forward, get_parallel, get_server_args, get_stream, ) from sglang.srt.utils import ( LazyValue, add_prefix, cpu_has_amx_support, is_cpu, is_cuda, is_hip, is_npu, make_layers, set_weight_attrs, ) logger = logging.getLogger(__name__) _is_cuda = is_cuda() _is_hip = is_hip() _is_npu = is_npu() _is_cpu = is_cpu() _is_amx_available = cpu_has_amx_support() if _is_npu: from sgl_kernel_npu.fla.utils import ( fused_qkvzba_split_reshape_cat as fused_qkvzba_split_reshape_cat_npu, ) from sgl_kernel_npu.norm.split_qkv_rmsnorm_rope import ( split_qkvgate_gemma_rmsnorm_rope, ) fused_qkvzba_split_reshape_cat = fused_qkvzba_split_reshape_cat_npu class Qwen3GatedDeltaNet(nn.Module): def __init__( self, config: Qwen3NextConfig, layer_id: int, quant_config: Optional[QuantizationConfig] = None, alt_stream: Optional[torch.cuda.Stream] = None, prefix: str = "", ) -> None: super().__init__() self.config = config self.attn_tp_rank = get_parallel().attn_tp_rank self.attn_tp_size = get_parallel().attn_tp_size self.hidden_size = config.hidden_size self.num_v_heads = ( config.linear_num_value_heads if not _is_cpu else config.linear_num_value_heads_cpu ) self.num_k_heads = ( config.linear_num_key_heads if not _is_cpu else config.linear_num_key_heads_cpu ) self.head_k_dim = config.linear_key_head_dim self.head_v_dim = config.linear_value_head_dim self.key_dim = self.head_k_dim * self.num_k_heads self.value_dim = self.head_v_dim * self.num_v_heads self.alt_stream = alt_stream self.conv_kernel_size = config.linear_conv_kernel_dim self.layer_id = layer_id self.activation = config.hidden_act self.output_gate_type = config.output_gate_type self.layer_norm_epsilon = config.rms_norm_eps self.conv_dim = self.key_dim * 2 + self.value_dim self.conv1d = ColumnParallelLinear( input_size=self.conv_kernel_size, output_size=self.conv_dim, bias=False, quant_config=None, tp_rank=self.attn_tp_rank, tp_size=self.attn_tp_size, prefix=add_prefix("conv1d", prefix), ) self.conv1d.weight.data = self.conv1d.weight.data.unsqueeze(1) # projection of the input hidden states self.in_proj_qkvz = self.create_qkvz_proj( hidden_size=self.hidden_size, key_dim=self.key_dim, value_dim=self.value_dim, quant_config=quant_config, prefix=add_prefix("in_proj_qkvz", prefix), tp_rank=self.attn_tp_rank, tp_size=self.attn_tp_size, ) self.in_proj_ba = MergedColumnParallelLinear( input_size=self.hidden_size, output_sizes=[self.num_v_heads] * 2, bias=False, quant_config=quant_config, prefix=add_prefix("in_proj_ba", prefix), tp_rank=self.attn_tp_rank, tp_size=self.attn_tp_size, ) # Override weight_loader for packed checkpoint format. # Must capture original_loader BEFORE overwriting. self._override_weight_loader( self.in_proj_qkvz, self._make_packed_weight_loader(self.in_proj_qkvz) ) self._override_weight_loader( self.in_proj_ba, self._make_packed_weight_loader(self.in_proj_ba) ) # Conv1d weight loader setup query_key_settings = (self.key_dim, 0, False) value_settings = (self.value_dim, 0, False) delattr(self.conv1d.weight, "weight_loader") set_weight_attrs( self.conv1d.weight, { "weight_loader": mamba_v2_sharded_weight_loader( [ query_key_settings, query_key_settings, value_settings, ], self.attn_tp_size, self.attn_tp_rank, ) }, ) self.dt_bias = nn.Parameter(torch.zeros(self.num_v_heads // self.attn_tp_size)) self.A_log = nn.Parameter( torch.zeros(self.num_v_heads // self.attn_tp_size, dtype=torch.float32) ) set_weight_attrs(self.A_log, {"weight_loader": sharded_weight_loader(0)}) set_weight_attrs(self.dt_bias, {"weight_loader": sharded_weight_loader(0)}) self.norm = ( RMSNormGated( self.head_v_dim, eps=self.layer_norm_epsilon, group_size=None, norm_before_gate=True, device=torch.get_device_module().current_device(), dtype=config.torch_dtype, **( {"activation": self.output_gate_type} if self.output_gate_type is not None else {} ), ) if check_cuda_graph_backend(Phase.PREFILL, Backend.TC_PIECEWISE) else FusedRMSNormGated( self.head_v_dim, eps=self.layer_norm_epsilon, activation=( self.output_gate_type if self.output_gate_type is not None else self.activation ), device=torch.get_device_module().current_device(), dtype=config.torch_dtype, ) ) self.out_proj = RowParallelLinear( self.value_dim, self.hidden_size, bias=False, quant_config=quant_config, input_is_parallel=True, reduce_results=False, tp_rank=self.attn_tp_rank, tp_size=self.attn_tp_size, prefix=add_prefix("out_proj", prefix), ) self.attn = RadixLinearAttention( layer_id=layer_id, num_q_heads=self.num_k_heads // self.attn_tp_size, num_k_heads=self.num_k_heads // self.attn_tp_size, num_v_heads=self.num_v_heads // self.attn_tp_size, head_q_dim=self.head_k_dim, head_k_dim=self.head_k_dim, head_v_dim=self.head_v_dim, conv_weights=self.conv1d.weight.squeeze(1), bias=self.conv1d.bias, activation=self.activation, A_log=self.A_log, dt_bias=self.dt_bias, ) @staticmethod def _override_weight_loader(module, new_loader): """Override weight_loader on a module's weight parameter. ModelWeightParameter exposes weight_loader as a read-only property backed by _weight_loader, while plain parameters store it as a regular attribute. This helper handles both cases.""" for attr_name in ( "weight", "weight_scale_inv", "weight_scale", "input_scale", "weight_offset", ): param = getattr(module, attr_name, None) if param is None: continue if hasattr(param, "_weight_loader"): param._weight_loader = new_loader else: param.weight_loader = new_loader @staticmethod def _make_packed_weight_loader(module): """Create a weight_loader that does contiguous TP slicing for fused (packed-format) checkpoint weights (shard_id=None), and delegates to the standard MergedColumnParallelLinear loader for split checkpoint weights (shard_id=int/tuple).""" original_loader = module.weight.weight_loader def weight_loader(param, loaded_weight, loaded_shard_id=None): if loaded_shard_id is None: # Fused checkpoint: weight is in packed (per-head-group) # format. Do contiguous TP slice like ColumnParallelLinear. output_dim = getattr(param, "output_dim", None) if output_dim is not None and module.tp_size > 1: shard_size = param.data.shape[output_dim] start_idx = module.tp_rank * shard_size if ( _is_cpu and _is_amx_available ) and start_idx + shard_size > loaded_weight.shape[output_dim]: shard_size = loaded_weight.shape[output_dim] - start_idx loaded_weight = loaded_weight.narrow( output_dim, start_idx, shard_size ) if _is_cpu and _is_amx_available: slices = tuple(slice(0, s) for s in loaded_weight.shape) param.data.zero_() param.data[slices].copy_(loaded_weight) else: assert param.data.shape == loaded_weight.shape, ( f"Shape mismatch: param {param.data.shape} vs " f"loaded {loaded_weight.shape}" ) param.data.copy_(loaded_weight) else: # Split checkpoint (int or tuple shard_id) → standard path original_loader(param, loaded_weight, loaded_shard_id) return weight_loader def create_qkvz_proj( self, hidden_size: int, key_dim: int, value_dim: int, quant_config: QuantizationConfig | None, prefix: str, tp_rank: Optional[int] = None, tp_size: Optional[int] = None, ) -> MergedColumnParallelLinear: return MergedColumnParallelLinear( input_size=hidden_size, output_sizes=[key_dim, key_dim, value_dim, value_dim], bias=False, quant_config=quant_config, prefix=prefix, tp_rank=tp_rank, tp_size=tp_size, ) def fix_query_key_value_ordering( self, mixed_qkvz: torch.Tensor, mixed_ba: torch.Tensor, ): """ Derives `query`, `key` and `value` tensors from `mixed_qkvzba`. """ new_tensor_shape_qkvz = mixed_qkvz.size()[:-1] + ( self.num_k_heads // self.attn_tp_size, ( self.head_k_dim + self.head_k_dim + (self.head_v_dim + self.head_v_dim) * self.num_v_heads // self.num_k_heads ), ) new_tensor_shape_ba = mixed_ba.size()[:-1] + ( self.num_k_heads // self.attn_tp_size, 2 * self.num_v_heads // self.num_k_heads, ) mixed_qkvz = mixed_qkvz.view(*new_tensor_shape_qkvz) mixed_ba = mixed_ba.view(*new_tensor_shape_ba) split_arg_list_qkvz = [ self.head_k_dim, self.head_k_dim, (self.num_v_heads // self.num_k_heads * self.head_v_dim), (self.num_v_heads // self.num_k_heads * self.head_v_dim), ] split_arg_list_ba = [ self.num_v_heads // self.num_k_heads, self.num_v_heads // self.num_k_heads, ] # [b, sq, ng, (hn + hn + np/ng * hn + np/ng + np/ng)] # --> [b, sq, ng, hn], [b, sq, ng, hn], [b, sq, ng, np/ng * hn], [b, sq, ng, np/ng * hn], [b, sq, ng, np/ng], [b, sq, ng, np/ng] query, key, value, z = torch.split(mixed_qkvz, split_arg_list_qkvz, dim=2) b, a = torch.split(mixed_ba, split_arg_list_ba, dim=2) # [b, sq, ng, np/ng * hn] -> [b, sq, np, hn] value = value.reshape(value.size(0), -1, self.head_v_dim) z = z.reshape(z.size(0), -1, self.head_v_dim) b = b.reshape(b.size(0), self.num_v_heads // self.attn_tp_size) a = a.reshape(a.size(0), self.num_v_heads // self.attn_tp_size) return query, key, value, z, b, a def _forward_input_proj(self, hidden_states: torch.Tensor): if ( _is_cpu or _is_npu or check_cuda_graph_backend(Phase.PREFILL, Backend.TC_PIECEWISE) ): DUAL_STREAM_TOKEN_THRESHOLD = 0 else: DUAL_STREAM_TOKEN_THRESHOLD = 1024 seq_len, _ = hidden_states.shape if ( self.alt_stream is not None and get_is_capture_mode() and seq_len < DUAL_STREAM_TOKEN_THRESHOLD ): current_stream = torch.cuda.current_stream() self.alt_stream.wait_stream(current_stream) projected_states_qkvz, _ = self.in_proj_qkvz(hidden_states) with torch.cuda.stream(self.alt_stream): projected_states_ba, _ = self.in_proj_ba(hidden_states) current_stream.wait_stream(self.alt_stream) else: projected_states_qkvz, _ = self.in_proj_qkvz(hidden_states) projected_states_ba, _ = self.in_proj_ba(hidden_states) return projected_states_qkvz, projected_states_ba def forward( self, hidden_states: torch.Tensor, forward_batch: ForwardBatch, ): projected_states_qkvz, projected_states_ba = self._forward_input_proj( hidden_states ) if self.num_v_heads // self.num_k_heads in [1, 2, 4] and not _is_cpu: mixed_qkv, z, b, a = fused_qkvzba_split_reshape_cat( projected_states_qkvz, projected_states_ba, triton.cdiv(self.num_k_heads, self.attn_tp_size), triton.cdiv(self.num_v_heads, self.attn_tp_size), self.head_k_dim, self.head_v_dim, ) elif _is_cpu and _is_amx_available: mixed_qkv, z, b, a = ( torch.ops.sgl_kernel.fused_qkvzba_split_reshape_cat_cpu( projected_states_qkvz, projected_states_ba, self.num_k_heads // self.attn_tp_size, self.num_v_heads // self.attn_tp_size, self.head_k_dim, self.head_v_dim, ) ) else: query, key, value, z, b, a = self.fix_query_key_value_ordering( projected_states_qkvz, projected_states_ba ) query, key, value = map( lambda x: x.reshape(x.shape[0], -1), (query, key, value) ) mixed_qkv = torch.cat((query, key, value), dim=-1) core_attn_out = self.attn( forward_batch, mixed_qkv=mixed_qkv, a=a, b=b, ) z_shape_og = z.shape # 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 core_attn_out.shape != z.shape: 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) output, _ = self.out_proj(core_attn_out) return output def _apply_qwen3_next_mlp( layer: nn.Module, hidden_states: torch.Tensor, residual: Optional[torch.Tensor], forward_batch: ForwardBatch, ) -> Tuple[torch.Tensor, Optional[torch.Tensor]]: hidden_states, residual = layer.layer_communicator.prepare_mlp( hidden_states, residual, forward_batch ) mlp_reduce_scatter = layer.layer_communicator.should_use_reduce_scatter( forward_batch ) fuse_mlp_allreduce = ( layer.layer_communicator.should_fuse_mlp_allreduce_with_next_layer( forward_batch ) ) with get_forward().scoped( fuse_mlp_allreduce=fuse_mlp_allreduce, mlp_reduce_scatter=mlp_reduce_scatter, ): if isinstance(layer.mlp, Qwen2MoeSparseMoeBlock): hidden_states = layer.mlp( hidden_states, forward_batch=forward_batch, ) else: hidden_states = layer.mlp(hidden_states) if fuse_mlp_allreduce: hidden_states._sglang_needs_allreduce_fusion = True else: hidden_states, residual = layer.layer_communicator.postprocess_layer( hidden_states, residual, forward_batch ) return hidden_states, residual class Qwen3HybridLinearDecoderLayer(nn.Module): def __init__( self, config: Qwen3NextConfig, layer_id: int, quant_config: Optional[QuantizationConfig] = None, prefix: str = "", alt_stream: Optional[torch.cuda.Stream] = None, is_nextn: bool = False, ) -> None: super().__init__() self.config = config self.linear_attn = Qwen3GatedDeltaNet( config, layer_id, quant_config, alt_stream, prefix ) # Qwen3Next all layers are sparse and have no nextn now self.is_layer_sparse = True is_previous_layer_sparse = True is_next_layer_sparse = True self.layer_id = layer_id self.layer_scatter_modes = LayerScatterModes.init_new( layer_id=layer_id, num_layers=config.num_hidden_layers, is_layer_sparse=self.is_layer_sparse, is_previous_layer_sparse=is_previous_layer_sparse, is_next_layer_sparse=is_next_layer_sparse, ) if self.is_layer_sparse: self.mlp = Qwen2MoeSparseMoeBlock( layer_id=layer_id, config=config, quant_config=quant_config, alt_stream=alt_stream, prefix=add_prefix("mlp", prefix.replace(".linear_attn", "")), is_nextn=is_nextn, support_shared_expert_fusion=True, enable_cuda_shared_expert_fusion=True, ) else: self.mlp = Qwen2MoeMLP( hidden_size=config.hidden_size, intermediate_size=config.intermediate_size, hidden_act=config.hidden_act, quant_config=quant_config, prefix=add_prefix("mlp", prefix.replace(".linear_attn", "")), ) self.input_layernorm = GemmaRMSNorm(config.hidden_size, eps=config.rms_norm_eps) self.post_attention_layernorm = GemmaRMSNorm( config.hidden_size, eps=config.rms_norm_eps ) self.layer_communicator = LayerCommunicator( layer_scatter_modes=self.layer_scatter_modes, input_layernorm=self.input_layernorm, post_attention_layernorm=self.post_attention_layernorm, allow_reduce_scatter=True, ) def forward( self, hidden_states: torch.Tensor, residual: Optional[torch.Tensor], captured_last_layer_outputs: Optional[list[torch.Tensor]] = None, **kwargs, ): forward_batch = kwargs.get("forward_batch", None) hidden_states, residual = ( self.layer_communicator.prepare_attn_and_capture_last_layer_outputs( hidden_states, residual, forward_batch, captured_last_layer_outputs=captured_last_layer_outputs, ) ) if not forward_batch.forward_mode.is_idle(): hidden_states = self.linear_attn( hidden_states, forward_batch, ) hidden_states, residual = _apply_qwen3_next_mlp( self, hidden_states, residual, forward_batch ) return hidden_states, residual class Qwen3HybridAttentionDecoderLayer(nn.Module): def __init__( self, config: Qwen3NextConfig, layer_id: int, quant_config: Optional[QuantizationConfig] = None, prefix: str = "", alt_stream: Optional[torch.cuda.Stream] = None, is_nextn: bool = False, ) -> None: super().__init__() self.config = config self.hidden_size = config.hidden_size self.attn_tp_rank = get_parallel().attn_tp_rank self.attn_tp_size = get_parallel().attn_tp_size self.total_num_heads = config.num_attention_heads assert self.total_num_heads % self.attn_tp_size == 0 self.num_heads = self.total_num_heads // self.attn_tp_size self.total_num_kv_heads = config.num_key_value_heads if self.total_num_kv_heads >= self.attn_tp_size: # Number of KV heads is greater than TP size, so we partition # the KV heads across multiple tensor parallel GPUs. assert self.total_num_kv_heads % self.attn_tp_size == 0 else: # Number of KV heads is less than TP size, so we replicate # the KV heads across multiple tensor parallel GPUs. assert self.attn_tp_size % self.total_num_kv_heads == 0 self.num_kv_heads = max(1, self.total_num_kv_heads // self.attn_tp_size) self.head_dim = config.head_dim or (self.hidden_size // self.num_heads) self.q_size = self.num_heads * self.head_dim self.kv_size = self.num_kv_heads * self.head_dim self.scaling = self.head_dim**-0.5 self.rope_theta = getattr(config, "rope_theta", 10000) self.max_position_embeddings = getattr(config, "max_position_embeddings", 8192) if "rope_parameters" in config: self.rope_scaling = getattr(config, "rope_parameters", None) else: self.rope_scaling = getattr(config, "rope_scaling", None) self.partial_rotary_factor = config.partial_rotary_factor self.layer_id = layer_id self.attn_output_gate = getattr(config, "attn_output_gate", True) if self.attn_output_gate: logger.warning_once("using attn output gate!") self.rotary_emb = get_rope( head_size=self.head_dim, rotary_dim=self.head_dim, max_position=self.max_position_embeddings, rope_scaling=self.rope_scaling, base=self.rope_theta, partial_rotary_factor=self.partial_rotary_factor, is_neox_style=True, dtype=torch.get_default_dtype(), # see impl of get_rope ) # qkv_proj is not quantized for fp4 self.qkv_proj = QKVParallelLinear( config.hidden_size, self.head_dim, self.total_num_heads * (1 + self.attn_output_gate), self.total_num_kv_heads, bias=False, quant_config=( quant_config if quant_config is not None and quant_config.get_name() != "modelopt_fp4" else None ), tp_rank=self.attn_tp_rank, tp_size=self.attn_tp_size, prefix=add_prefix("qkv_proj", prefix), ) self.o_proj = RowParallelLinear( self.total_num_heads * self.head_dim, config.hidden_size, bias=False, quant_config=quant_config, reduce_results=False, tp_rank=self.attn_tp_rank, tp_size=self.attn_tp_size, prefix=add_prefix("o_proj", prefix), ) self.attn = RadixAttention( self.num_heads, self.head_dim, self.scaling, num_kv_heads=self.num_kv_heads, layer_id=layer_id, quant_config=quant_config, prefix=f"{prefix}.attn", ) # Qwen3Next all layers are sparse and have no nextn now self.is_layer_sparse = True is_previous_layer_sparse = True is_next_layer_sparse = True self.layer_scatter_modes = LayerScatterModes.init_new( layer_id=layer_id, num_layers=config.num_hidden_layers, is_layer_sparse=self.is_layer_sparse, is_previous_layer_sparse=is_previous_layer_sparse, is_next_layer_sparse=is_next_layer_sparse, ) if self.is_layer_sparse: self.mlp = Qwen2MoeSparseMoeBlock( layer_id=layer_id, config=config, quant_config=quant_config, alt_stream=alt_stream, prefix=add_prefix("mlp", prefix.replace(".self_attn", "")), is_nextn=is_nextn, support_shared_expert_fusion=True, enable_cuda_shared_expert_fusion=True, ) else: self.mlp = Qwen2MoeMLP( hidden_size=config.hidden_size, intermediate_size=config.intermediate_size, hidden_act=config.hidden_act, quant_config=quant_config, prefix=add_prefix("mlp", prefix.replace(".self_attn", "")), ) self.input_layernorm = GemmaRMSNorm(config.hidden_size, eps=config.rms_norm_eps) self.post_attention_layernorm = GemmaRMSNorm( config.hidden_size, eps=config.rms_norm_eps ) self.q_norm = GemmaRMSNorm(self.head_dim, eps=config.rms_norm_eps) self.k_norm = GemmaRMSNorm(self.head_dim, eps=config.rms_norm_eps) self.layer_communicator = LayerCommunicator( layer_scatter_modes=self.layer_scatter_modes, input_layernorm=self.input_layernorm, post_attention_layernorm=self.post_attention_layernorm, allow_reduce_scatter=True, ) self.alt_stream = alt_stream def _apply_qk_norm( self, q: torch.Tensor, k: torch.Tensor ) -> Tuple[torch.Tensor, torch.Tensor]: # overlap qk norm if self.alt_stream is not None and get_is_capture_mode(): current_stream = torch.cuda.current_stream() self.alt_stream.wait_stream(current_stream) q_by_head = q.reshape(-1, self.head_dim) q_by_head = self.q_norm(q_by_head) with torch.cuda.stream(self.alt_stream): k_by_head = k.reshape(-1, self.head_dim) k_by_head = self.k_norm(k_by_head) current_stream.wait_stream(self.alt_stream) else: q_by_head = q.reshape(-1, self.head_dim) q_by_head = self.q_norm(q_by_head) k_by_head = k.reshape(-1, self.head_dim) k_by_head = self.k_norm(k_by_head) q = q_by_head.view(q.shape) k = k_by_head.view(k.shape) return q, k def forward_prepare_native(self, positions, hidden_states): qkv, _ = self.qkv_proj(hidden_states) if self.attn_output_gate: q_gate, k, v = qkv.split( [self.q_size * 2, self.kv_size, self.kv_size], dim=-1 ) orig_shape = q_gate.shape[:-1] q_gate = q_gate.view(*orig_shape, self.num_heads, -1) q, gate = torch.chunk(q_gate, 2, dim=-1) q = q.reshape(*orig_shape, -1) gate = gate.reshape(*orig_shape, -1) else: q, k, v = qkv.split([self.q_size, self.kv_size, self.kv_size], dim=-1) gate = None q, k = self._apply_qk_norm(q, k) q, k = self.rotary_emb(positions, q, k) return q, k, v, gate def forward_prepare_npu(self, positions, hidden_states, forward_batch): qkv, _ = self.qkv_proj(hidden_states) # Calculate first full attention layer ID based on config if self.attn.layer_id == (self.config.full_attention_interval - 1): self.rotary_emb.get_cos_sin_with_position(positions) q, k, v, gate = split_qkvgate_gemma_rmsnorm_rope( qkv, self.rotary_emb.position_sin, self.rotary_emb.position_cos, self.q_size, self.kv_size, self.head_dim, int(self.head_dim * self.partial_rotary_factor), eps=self.q_norm.variance_epsilon, q_weight=self.q_norm.weight, k_weight=self.k_norm.weight, ) return q, k, v, gate def self_attention( self, positions: torch.Tensor, hidden_states: torch.Tensor, forward_batch: ForwardBatch, ) -> torch.Tensor: """Full attention forward pass.""" if ( not _is_npu or forward_batch.forward_mode.is_extend_or_draft_extend_or_mixed() or not self.attn_output_gate ): q, k, v, gate = self.forward_prepare_native( positions=positions, hidden_states=hidden_states, ) else: q, k, v, gate = self.forward_prepare_npu( positions=positions, hidden_states=hidden_states, forward_batch=forward_batch, ) attn_output = self.attn(q, k, v, forward_batch) if self.attn_output_gate: if _is_hip: from sglang.jit_kernel.triton.sigmoid_gate_mul import ( sigmoid_gate_mul, ) attn_output = sigmoid_gate_mul(attn_output, gate) else: gate = torch.sigmoid(gate) attn_output = attn_output * gate output, _ = self.o_proj(attn_output) return output def forward( self, positions: torch.Tensor, hidden_states: torch.Tensor, residual: Optional[torch.Tensor], forward_batch: ForwardBatch, captured_last_layer_outputs: Optional[list[torch.Tensor]] = None, **kwargs: Any, ): hidden_states, residual = ( self.layer_communicator.prepare_attn_and_capture_last_layer_outputs( hidden_states, residual, forward_batch, captured_last_layer_outputs=captured_last_layer_outputs, ) ) if not forward_batch.forward_mode.is_idle(): hidden_states = self.self_attention( positions=positions, hidden_states=hidden_states, forward_batch=forward_batch, ) hidden_states, residual = _apply_qwen3_next_mlp( self, hidden_states, residual, forward_batch ) return hidden_states, residual ALL_DECODER_LAYER_TYPES = { "attention": Qwen3HybridAttentionDecoderLayer, "linear_attention": Qwen3HybridLinearDecoderLayer, } class Qwen3NextModel(nn.Module): def __init__( self, config: Qwen3NextConfig, quant_config: Optional[QuantizationConfig] = None, prefix: str = "", is_nextn: bool = False, ) -> None: super().__init__() self.config = config alt_stream = get_stream("alt") if _is_cuda else None self.embed_tokens = VocabParallelEmbedding( config.vocab_size, config.hidden_size, org_num_embeddings=config.vocab_size, use_attn_tp_group=is_dp_attention_enabled(), ) def get_layer(idx: int, prefix: str): layer_class = ALL_DECODER_LAYER_TYPES[config.layers_block_type[idx]] if config.layers_block_type[idx] == "attention": prefix = add_prefix("self_attn", prefix) else: prefix = add_prefix("linear_attn", prefix) return layer_class( config, idx, quant_config=quant_config, prefix=prefix, alt_stream=alt_stream, is_nextn=is_nextn, ) self.layers = make_layers( config.num_hidden_layers, get_layer, prefix=f"{prefix}.layers" ) self.norm = GemmaRMSNorm(config.hidden_size, eps=config.rms_norm_eps) self.infer_count = 0 # For EAGLE3 support self.layers_to_capture = [] def set_eagle3_layers_to_capture(self, layers_to_capture: list[int]): self.layers_to_capture = layers_to_capture for layer_id in self.layers_to_capture: setattr(self.layers[layer_id], "_is_layer_to_capture", True) def set_dflash_layers_to_capture(self, layers_to_capture: list[int]): self.layers_to_capture = layers_to_capture for layer_id in self.layers_to_capture: setattr(self.layers[layer_id], "_is_layer_to_capture", True) def forward( self, input_ids: torch.Tensor, positions: torch.Tensor, forward_batch: ForwardBatch, # mamba_cache_params: MambaCacheParams, inputs_embeds: Optional[torch.Tensor] = None, ) -> torch.Tensor: # pass a sequence index tensor, that is required for # proper continuous batching computation including # chunked prefill if inputs_embeds is not None: hidden_states = inputs_embeds else: hidden_states = self.embed_tokens(input_ids) residual = None aux_hidden_states = [] for i in range(len(self.layers)): layer = self.layers[i] with get_global_expert_distribution_recorder().with_current_layer(i): hidden_states, residual = layer( layer_id=i, positions=positions, hidden_states=hidden_states, residual=residual, forward_batch=forward_batch, captured_last_layer_outputs=( aux_hidden_states if getattr(layer, "_is_layer_to_capture", False) else None ), ) if not forward_batch.forward_mode.is_idle(): if residual is None: hidden_states = self.norm(hidden_states) else: hidden_states, _ = self.norm(hidden_states, residual) if len(aux_hidden_states) == 0: return hidden_states return hidden_states, aux_hidden_states class HybridLayerType(enum.Enum): full_attention = "attention" swa_attention = "swa_attention" linear_attention = "linear_attention" mamba2 = "mamba" class Qwen3NextForCausalLM(nn.Module): fall_back_to_pt_during_load = False # Map fused module names to their checkpoint (unfused) counterparts. # This is needed so the quantization exclusion logic can match # checkpoint-style names (e.g. "q_proj") against the fused sglang # module names (e.g. "qkv_proj"). packed_modules_mapping = { "qkv_proj": ["q_proj", "k_proj", "v_proj"], "gate_up_proj": ["gate_proj", "up_proj"], } def __init__( self, config: Qwen3NextConfig, quant_config: Optional[QuantizationConfig] = None, prefix: str = "", ) -> None: super().__init__() self.config = config self.pp_group = get_pp_group() assert self.pp_group.is_first_rank and self.pp_group.is_last_rank # The quant config's packed_modules_mapping may be None if it wasn't # in the checkpoint config. The base class (QuantizationConfig) intends # for models to set this. We need it so is_layer_skipped can unfuse # "qkv_proj" into ["q_proj","k_proj","v_proj"] when checking exclusions. if quant_config is not None and hasattr(quant_config, "packed_modules_mapping"): quant_config.packed_modules_mapping = self.packed_modules_mapping self.quant_config = quant_config self.model = Qwen3NextModel( config, quant_config, prefix=add_prefix("model", prefix) ) self.lm_head = ParallelLMHead( config.vocab_size, config.hidden_size, quant_config=quant_config, org_num_embeddings=config.vocab_size, prefix=add_prefix("lm_head", prefix), use_attn_tp_group=get_server_args().enable_dp_lm_head, ) self.logits_processor = LogitsProcessor(config) # For EAGLE3 support self.capture_aux_hidden_states = False self.num_fused_shared_experts = self._get_num_fused_shared_experts() if self.num_fused_shared_experts > 1: raise ValueError( "Qwen3-Next shared expert fusion currently supports exactly one " "shared expert because checkpoint weight remapping maps it into " "a single fused MoE expert slot." ) self.enable_shared_expert_fusion = self.num_fused_shared_experts > 0 self._routed_experts_weights_of_layer = LazyValue( lambda: { layer_id: layer.mlp.get_moe_weights() for layer_id, layer in enumerate(self.model.layers) if isinstance(layer.mlp, Qwen2MoeSparseMoeBlock) } ) @property def routed_experts_weights_of_layer(self): return self._routed_experts_weights_of_layer.value def _get_num_fused_shared_experts(self) -> int: if not hasattr(self.model, "layers"): return 0 for layer in self.model.layers: if isinstance(layer.mlp, Qwen2MoeSparseMoeBlock): return layer.mlp.num_fused_shared_experts return 0 @torch.no_grad() def forward( self, input_ids: torch.Tensor, positions: torch.Tensor, forward_batch: ForwardBatch, inputs_embeds: Optional[torch.Tensor] = None, **kwargs, ): hidden_states = self.model(input_ids, positions, forward_batch, inputs_embeds) aux_hidden_states = None if self.capture_aux_hidden_states: hidden_states, aux_hidden_states = hidden_states return self.logits_processor( input_ids, hidden_states, self.lm_head, forward_batch, aux_hidden_states ) def get_embed_and_head(self): return self.model.embed_tokens.weight, self.lm_head.weight def get_input_embeddings(self) -> nn.Embedding: return self.model.embed_tokens def set_embed_and_head(self, embed, head): del self.model.embed_tokens.weight del self.lm_head.weight self.model.embed_tokens.weight = embed self.lm_head.weight = head torch.cuda.empty_cache() torch.cuda.synchronize() def get_embed(self): return self.model.embed_tokens.weight def set_embed(self, embed): # NOTE: If draft hidden size != target hidden size, the embed weight cannot be shared for EAGLE3 if ( hasattr(self.config, "target_hidden_size") and self.config.target_hidden_size != self.config.hidden_size ): return del self.model.embed_tokens.weight self.model.embed_tokens.weight = embed torch.cuda.empty_cache() torch.cuda.synchronize() def load_weights( self, weights: Iterable[Tuple[str, torch.Tensor]], is_mtp: bool = False ) -> Set[str]: stacked_params_mapping = [ # (param_name, shard_name, shard_id) # self attention ("qkv_proj", "q_proj", "q"), ("qkv_proj", "k_proj", "k"), ("qkv_proj", "v_proj", "v"), # mlp ("gate_up_proj", "gate_proj", 0), ("gate_up_proj", "up_proj", 1), # GDN ("in_proj_qkvz.", "in_proj_qkv.", (0, 1, 2)), ("in_proj_qkvz.", "in_proj_z.", 3), ("in_proj_ba.", "in_proj_b.", 0), ("in_proj_ba.", "in_proj_a.", 1), ] # Params for weights, fp8 weight scales, fp8 activation scales # (param_name, weight_name, expert_id, shard_id) expert_params_mapping = FusedMoE.make_expert_params_mapping( ckpt_gate_proj_name="gate_proj", ckpt_down_proj_name="down_proj", ckpt_up_proj_name="up_proj", num_experts=( self.config.num_experts if not self.enable_shared_expert_fusion else self.config.num_experts + self.num_fused_shared_experts ), ) params_dict = dict(self.named_parameters()) loaded_params: Set[str] = set() for name, loaded_weight in weights: if is_mtp: if "mtp" not in name: continue if name in [ "mtp.fc.weight", "mtp.pre_fc_norm_embedding.weight", "mtp.pre_fc_norm_hidden.weight", ]: name = name.replace("mtp.", "") else: name = name.replace("mtp", "model") if not is_mtp and "mtp" in name: continue if "rotary_emb.inv_freq" in name: continue if ".self_attn." in name: name = name.replace(".self_attn", "") if self.enable_shared_expert_fusion and "mlp.shared_expert." in name: name = name.replace( "mlp.shared_expert.", f"mlp.experts.{self.config.num_experts}.", ) # Remap modelopt FP8 KV cache scale names: # checkpoint: k_proj.k_scale / v_proj.v_scale # model: attn.k_scale / attn.v_scale if name.endswith(".k_proj.k_scale"): name = name.replace(".k_proj.k_scale", ".attn.k_scale") elif name.endswith(".v_proj.v_scale"): name = name.replace(".v_proj.v_scale", ".attn.v_scale") for param_name, weight_name, shard_id in stacked_params_mapping: if weight_name not in name: continue # TODO(fix mtp loading) if "mlp.experts" in name: continue replaced_name = name.replace(weight_name, param_name) # Skip loading extra bias for GPTQ models. if replaced_name.endswith(".bias") and replaced_name not in params_dict: continue # Skip layers on other devices. # if is_pp_missing_parameter(name, self): # continue if replaced_name not in params_dict: continue name = replaced_name param = params_dict[name] weight_loader = getattr(param, "weight_loader") weight_loader(param, loaded_weight, shard_id) break else: for mapping in expert_params_mapping: param_name, weight_name, expert_id, shard_id = mapping if weight_name not in name: continue replaced_name = name.replace(weight_name, param_name) # Skip layers on other devices. # if is_pp_missing_parameter(name, self): # continue # Skip loading extra bias for GPTQ models. if ( replaced_name.endswith(".bias") or replaced_name.endswith("_bias") ) and replaced_name not in params_dict: continue name = replaced_name param = params_dict[name] weight_loader = getattr(param, "weight_loader") weight_loader( param, loaded_weight, name, shard_id=shard_id, expert_id=expert_id, ) break else: # Skip loading extra bias for GPTQ models. if name.endswith(".bias") and name not in params_dict: continue # if is_pp_missing_parameter(name, self): # continue if name.endswith("_scale") and name not in params_dict: assert ( abs(loaded_weight.item() - 1.0) < 1e-6 ), f"Expected 1.0, got {loaded_weight.item()} in skipped {name}" continue param = params_dict[name] weight_loader = getattr( param, "weight_loader", default_weight_loader ) weight_loader(param, loaded_weight) loaded_params.add(name) return loaded_params @classmethod def get_model_config_for_expert_location(cls, config): return ModelConfigForExpertLocation( num_layers=config.num_hidden_layers, num_logical_experts=config.num_experts, num_groups=None, ) def set_eagle3_layers_to_capture(self, layer_ids: Optional[list[int]] = None): if not self.pp_group.is_last_rank: return self.capture_aux_hidden_states = True if layer_ids is None: num_layers = self.config.num_hidden_layers self.model.set_eagle3_layers_to_capture( [ 2, num_layers // 2, num_layers - 3, ] ) # Specific layers for EAGLE3 support else: self.model.set_eagle3_layers_to_capture([val + 1 for val in layer_ids]) def set_dflash_layers_to_capture(self, layer_ids: list[int]): if not self.pp_group.is_last_rank: return if layer_ids is None: raise ValueError( "DFLASH requires explicit layer_ids for aux hidden capture." ) self.capture_aux_hidden_states = True self.model.set_dflash_layers_to_capture([val + 1 for val in layer_ids]) EntryClass = Qwen3NextForCausalLM