[Model] Add Qwen3 and Qwen3MoE (vllm-project#15289)

Signed-off-by: YamPengLi <[email protected]>
Co-authored-by: Cyrus Leung <[email protected]>

Author: 2 people

docs/source/models/supported_models.md

@@ -478,6 +478,16 @@ See [this page](#generative-models) for more information on how to use generativ
   * `Qwen/Qwen1.5-MoE-A2.7B`, `Qwen/Qwen1.5-MoE-A2.7B-Chat`, etc.
   *
   * ✅︎
+- * `Qwen3ForCausalLM`
+  * Qwen3
+  * `Qwen/Qwen3-8B`, etc.
+  * ✅︎
+  * ✅︎
+- * `Qwen3MoeForCausalLM`
+  * Qwen3MoE
+  * `Qwen/Qwen3-MoE-15B-A2B`, etc.
+  * ✅︎
+  * ✅︎
 - * `StableLmForCausalLM`
   * StableLM
   * `stabilityai/stablelm-3b-4e1t`, `stabilityai/stablelm-base-alpha-7b-v2`, etc.

tests/models/registry.py

@@ -202,6 +202,16 @@ def check_available_online(
     "Qwen2ForCausalLM": _HfExamplesInfo("Qwen/Qwen2-7B-Instruct",
                                         extras={"2.5": "Qwen/Qwen2.5-7B-Instruct"}), # noqa: E501
     "Qwen2MoeForCausalLM": _HfExamplesInfo("Qwen/Qwen1.5-MoE-A2.7B-Chat"),
+    "Qwen3ForCausalLM": _HfExamplesInfo(
+        "Qwen/Qwen3-8B",
+        is_available_online=False,
+        min_transformers_version="4.51"
+    ),
+    "Qwen3MoeForCausalLM": _HfExamplesInfo(
+        "Qwen/Qwen3-MoE-15B-A2B",
+        is_available_online=False,
+        min_transformers_version="4.51"
+    ),
     "RWForCausalLM": _HfExamplesInfo("tiiuae/falcon-40b",
                                      is_available_online=False),
     "StableLMEpochForCausalLM": _HfExamplesInfo("stabilityai/stablelm-zephyr-3b",  # noqa: E501

vllm/model_executor/models/qwen2.py

@@ -263,7 +263,11 @@ def forward(
     })
 class Qwen2Model(nn.Module):
 
-    def __init__(self, *, vllm_config: VllmConfig, prefix: str = ""):
+    def __init__(self,
+                 *,
+                 vllm_config: VllmConfig,
+                 prefix: str = "",
+                 decoder_layer_type: type[nn.Module] = Qwen2DecoderLayer):
         super().__init__()
 
         config = vllm_config.model_config.hf_config
@@ -297,12 +301,14 @@ def __init__(self, *, vllm_config: VllmConfig, prefix: str = ""):
         else:
             self.embed_tokens = PPMissingLayer()
 
+        # Use the provided decoder layer type or default to Qwen2DecoderLayer
+        decoder_layer_type = decoder_layer_type or Qwen2DecoderLayer
         self.start_layer, self.end_layer, self.layers = make_layers(
             config.num_hidden_layers,
-            lambda prefix: Qwen2DecoderLayer(config=config,
-                                             cache_config=cache_config,
-                                             quant_config=quant_config,
-                                             prefix=prefix),
+            lambda prefix: decoder_layer_type(config=config,
+                                              cache_config=cache_config,
+                                              quant_config=quant_config,
+                                              prefix=prefix),
             prefix=f"{prefix}.layers",
         )
 

vllm/model_executor/models/qwen3.py

@@ -0,0 +1,329 @@
+# SPDX-License-Identifier: Apache-2.0
+
+# Copyright 2024 The Qwen team.
+# Copyright 2023 The vLLM team.
+# Copyright 2022 EleutherAI and the HuggingFace Inc. team. All rights reserved.
+#
+# This code is based on EleutherAI's GPT-NeoX library and the GPT-NeoX
+# and OPT implementations in this library. It has been modified from its
+# original forms to accommodate minor architectural differences compared
+# to GPT-NeoX and OPT used by the Meta AI team that trained the model.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Inference-only Qwen3 model compatible with HuggingFace weights."""
+from typing import Iterable, Optional, Set, Tuple, Union
+
+import torch
+from torch import nn
+from transformers import Qwen3Config
+
+from vllm.attention import Attention, AttentionType
+from vllm.compilation.decorators import support_torch_compile
+from vllm.config import CacheConfig, VllmConfig
+from vllm.distributed import get_pp_group, get_tensor_model_parallel_world_size
+from vllm.logger import init_logger
+from vllm.model_executor.layers.layernorm import RMSNorm
+from vllm.model_executor.layers.linear import (QKVParallelLinear,
+                                               RowParallelLinear)
+from vllm.model_executor.layers.logits_processor import LogitsProcessor
+from vllm.model_executor.layers.quantization import QuantizationConfig
+from vllm.model_executor.layers.rotary_embedding import get_rope
+from vllm.model_executor.layers.sampler import SamplerOutput, get_sampler
+from vllm.model_executor.layers.vocab_parallel_embedding import ParallelLMHead
+from vllm.model_executor.sampling_metadata import SamplingMetadata
+from vllm.sequence import IntermediateTensors
+
+from .interfaces import SupportsLoRA, SupportsPP
+from .qwen2 import Qwen2MLP as Qwen3MLP
+from .qwen2 import Qwen2Model
+from .utils import AutoWeightsLoader, PPMissingLayer, maybe_prefix
+
+logger = init_logger(__name__)
+
+
+class Qwen3Attention(nn.Module):
+
+    def __init__(self,
+                 hidden_size: int,
+                 num_heads: int,
+                 num_kv_heads: int,
+                 max_position: int = 4096 * 32,
+                 head_dim: Optional[int] = None,
+                 rms_norm_eps: float = 1e-06,
+                 qkv_bias: bool = False,
+                 rope_theta: float = 10000,
+                 cache_config: Optional[CacheConfig] = None,
+                 quant_config: Optional[QuantizationConfig] = None,
+                 rope_scaling: Optional[Tuple] = None,
+                 prefix: str = "",
+                 attn_type: str = AttentionType.DECODER) -> None:
+        super().__init__()
+        self.hidden_size = hidden_size
+        tp_size = get_tensor_model_parallel_world_size()
+        self.total_num_heads = num_heads
+        assert self.total_num_heads % tp_size == 0
+        self.num_heads = self.total_num_heads // tp_size
+        self.total_num_kv_heads = num_kv_heads
+        if self.total_num_kv_heads >= 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 % 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 tp_size % self.total_num_kv_heads == 0
+        self.num_kv_heads = max(1, self.total_num_kv_heads // tp_size)
+        self.head_dim = head_dim or hidden_size // self.total_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 = rope_theta
+
+        self.qkv_proj = QKVParallelLinear(
+            hidden_size,
+            self.head_dim,
+            self.total_num_heads,
+            self.total_num_kv_heads,
+            bias=qkv_bias,
+            quant_config=quant_config,
+            prefix=f"{prefix}.qkv_proj",
+        )
+        self.o_proj = RowParallelLinear(
+            self.total_num_heads * self.head_dim,
+            hidden_size,
+            bias=False,
+            quant_config=quant_config,
+            prefix=f"{prefix}.o_proj",
+        )
+
+        self.rotary_emb = get_rope(
+            self.head_dim,
+            rotary_dim=self.head_dim,
+            max_position=max_position,
+            base=self.rope_theta,
+            rope_scaling=rope_scaling,
+        )
+        self.attn = Attention(self.num_heads,
+                              self.head_dim,
+                              self.scaling,
+                              num_kv_heads=self.num_kv_heads,
+                              cache_config=cache_config,
+                              quant_config=quant_config,
+                              prefix=f"{prefix}.attn",
+                              attn_type=attn_type)
+        self.q_norm = RMSNorm(self.head_dim, eps=rms_norm_eps)
+        self.k_norm = RMSNorm(self.head_dim, eps=rms_norm_eps)
+
+    def forward(
+        self,
+        positions: torch.Tensor,
+        hidden_states: torch.Tensor,
+    ) -> torch.Tensor:
+        qkv, _ = self.qkv_proj(hidden_states)
+        q, k, v = qkv.split([self.q_size, self.kv_size, self.kv_size], dim=-1)
+        # Add qk-norm
+        q_by_head = q.view(*q.shape[:-1], q.shape[-1] // self.head_dim,
+                           self.head_dim)
+        q_by_head = self.q_norm.forward_native(q_by_head)
+        q = q_by_head.view(q.shape)
+        k_by_head = k.view(*k.shape[:-1], k.shape[-1] // self.head_dim,
+                           self.head_dim)
+        k_by_head = self.k_norm.forward_native(k_by_head)
+        k = k_by_head.view(k.shape)
+        q, k = self.rotary_emb(positions, q, k)
+        attn_output = self.attn(q, k, v)
+        output, _ = self.o_proj(attn_output)
+        return output
+
+
+class Qwen3DecoderLayer(nn.Module):
+
+    def __init__(
+        self,
+        config: Qwen3Config,
+        cache_config: Optional[CacheConfig] = None,
+        quant_config: Optional[QuantizationConfig] = None,
+        prefix: str = "",
+    ) -> None:
+        super().__init__()
+        self.hidden_size = config.hidden_size
+        # Requires transformers > 4.32.0
+        rope_theta = getattr(config, "rope_theta", 1000000)
+        rope_scaling = getattr(config, "rope_scaling", None)
+
+        # By default, Qwen3 uses causal attention as it is a decoder-only model.
+        # You can override the HF config with `is_causal=False` to enable
+        # bidirectional attention, which is used in some embedding models
+        # (e.g. Alibaba-NLP/gte-Qwen3-7B-instruct)
+        if getattr(config, "is_causal", True):
+            attn_type = AttentionType.DECODER
+        else:
+            attn_type = AttentionType.ENCODER_ONLY
+
+        self.self_attn = Qwen3Attention(
+            hidden_size=self.hidden_size,
+            num_heads=config.num_attention_heads,
+            max_position=config.max_position_embeddings,
+            num_kv_heads=config.num_key_value_heads,
+            rope_theta=rope_theta,
+            rms_norm_eps=config.rms_norm_eps,
+            qkv_bias=getattr(config, 'attention_bias', False),
+            head_dim=getattr(config, 'head_dim', None),
+            cache_config=cache_config,
+            quant_config=quant_config,
+            rope_scaling=rope_scaling,
+            prefix=f"{prefix}.self_attn",
+            attn_type=attn_type,
+        )
+        self.mlp = Qwen3MLP(
+            hidden_size=self.hidden_size,
+            intermediate_size=config.intermediate_size,
+            hidden_act=config.hidden_act,
+            quant_config=quant_config,
+            prefix=f"{prefix}.mlp",
+        )
+        self.input_layernorm = RMSNorm(config.hidden_size,
+                                       eps=config.rms_norm_eps)
+        self.post_attention_layernorm = RMSNorm(config.hidden_size,
+                                                eps=config.rms_norm_eps)
+
+    def forward(
+        self,
+        positions: torch.Tensor,
+        hidden_states: torch.Tensor,
+        residual: Optional[torch.Tensor],
+    ) -> Tuple[torch.Tensor, torch.Tensor]:
+        # Self Attention
+        if residual is None:
+            residual = hidden_states
+            hidden_states = self.input_layernorm(hidden_states)
+        else:
+            hidden_states, residual = self.input_layernorm(
+                hidden_states, residual)
+        hidden_states = self.self_attn(
+            positions=positions,
+            hidden_states=hidden_states,
+        )
+
+        # Fully Connected
+        hidden_states, residual = self.post_attention_layernorm(
+            hidden_states, residual)
+        hidden_states = self.mlp(hidden_states)
+        return hidden_states, residual
+
+
+ALL_DECODER_LAYER_TYPES = {
+    "attention": Qwen3DecoderLayer,
+}
+
+
+@support_torch_compile(
+    dynamic_arg_dims={
+        "input_ids": 0,
+        # positions is of shape (3, seq_len) if mrope is enabled for qwen2-vl,
+        # otherwise (seq_len, ).
+        "positions": -1,
+        "intermediate_tensors": 0,
+        "inputs_embeds": 0,
+    })
+class Qwen3Model(Qwen2Model):
+
+    def __init__(self, *, vllm_config: VllmConfig, prefix: str = ""):
+        super().__init__(vllm_config=vllm_config,
+                         prefix=prefix,
+                         decoder_layer_type=Qwen3DecoderLayer)
+
+
+class Qwen3ForCausalLM(nn.Module, SupportsLoRA, SupportsPP):
+    packed_modules_mapping = {
+        "qkv_proj": [
+            "q_proj",
+            "k_proj",
+            "v_proj",
+        ],
+        "gate_up_proj": [
+            "gate_proj",
+            "up_proj",
+        ],
+    }
+
+    def __init__(self, *, vllm_config: VllmConfig, prefix: str = ""):
+        super().__init__()
+        config = vllm_config.model_config.hf_config
+        quant_config = vllm_config.quant_config
+        lora_config = vllm_config.lora_config
+
+        self.config = config
+        self.lora_config = lora_config
+
+        self.quant_config = quant_config
+        self.model = Qwen3Model(vllm_config=vllm_config,
+                                prefix=maybe_prefix(prefix, "model"))
+
+        if get_pp_group().is_last_rank:
+            if config.tie_word_embeddings:
+                self.lm_head = self.model.embed_tokens
+            else:
+                self.lm_head = ParallelLMHead(config.vocab_size,
+                                              config.hidden_size,
+                                              quant_config=quant_config,
+                                              prefix=maybe_prefix(
+                                                  prefix, "lm_head"))
+        else:
+            self.lm_head = PPMissingLayer()
+
+        self.logits_processor = LogitsProcessor(config.vocab_size)
+        self.sampler = get_sampler()
+
+        self.make_empty_intermediate_tensors = (
+            self.model.make_empty_intermediate_tensors)
+
+    def get_input_embeddings(self, input_ids: torch.Tensor) -> torch.Tensor:
+        return self.model.get_input_embeddings(input_ids)
+
+    def forward(
+        self,
+        input_ids: torch.Tensor,
+        positions: torch.Tensor,
+        intermediate_tensors: Optional[IntermediateTensors] = None,
+        inputs_embeds: Optional[torch.Tensor] = None,
+    ) -> Union[torch.Tensor, IntermediateTensors]:
+        hidden_states = self.model(input_ids, positions, intermediate_tensors,
+                                   inputs_embeds)
+        return hidden_states
+
+    def compute_logits(
+        self,
+        hidden_states: torch.Tensor,
+        sampling_metadata: SamplingMetadata,
+    ) -> Optional[torch.Tensor]:
+        logits = self.logits_processor(self.lm_head, hidden_states,
+                                       sampling_metadata)
+        return logits
+
+    def sample(
+        self,
+        logits: torch.Tensor,
+        sampling_metadata: SamplingMetadata,
+    ) -> Optional[SamplerOutput]:
+        next_tokens = self.sampler(logits, sampling_metadata)
+        return next_tokens
+
+    def load_weights(self, weights: Iterable[Tuple[str,
+                                                   torch.Tensor]]) -> Set[str]:
+        loader = AutoWeightsLoader(
+            self,
+            skip_prefixes=(["lm_head."]
+                           if self.config.tie_word_embeddings else None),
+        )
+        return loader.load_weights(weights)