mirror of
https://github.com/PaddlePaddle/FastDeploy.git
synced 2026-04-23 08:21:53 +08:00
[Models] Add Qwen3-VL Moe Model Support (#5913)
* [Model] add Qwen3vl moe model support * [Model] add Qwen3vl moe model support remove log * [Model] add Qwen3vl moe model support unittest
This commit is contained in:
CSWYF3634076
@@ -1,3 +1,19 @@
|
||||
"""
|
||||
# Copyright (c) 2026 PaddlePaddle Authors. All Rights Reserved.
|
||||
#
|
||||
# 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.
|
||||
"""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
import re
|
||||
|
||||
@@ -0,0 +1,278 @@
|
||||
"""
|
||||
# Copyright (c) 2026 PaddlePaddle Authors. All Rights Reserved.
|
||||
#
|
||||
# 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.
|
||||
"""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
import re
|
||||
from typing import List, Optional
|
||||
|
||||
import paddle
|
||||
from paddle import nn
|
||||
from paddleformers.utils.log import logger
|
||||
|
||||
from fastdeploy.config import FDConfig
|
||||
from fastdeploy.model_executor.forward_meta import ForwardMeta
|
||||
from fastdeploy.model_executor.layers.embeddings import VocabParallelEmbedding
|
||||
from fastdeploy.model_executor.layers.moe.moe import FusedMoE
|
||||
from fastdeploy.model_executor.layers.normalization import RMSNorm
|
||||
from fastdeploy.model_executor.models.model_base import ModelCategory, ModelRegistry
|
||||
from fastdeploy.model_executor.models.qwen3_vl.qwen3_vl import (
|
||||
Qwen3VLForConditionalGeneration,
|
||||
)
|
||||
from fastdeploy.model_executor.models.qwen3moe import (
|
||||
Qwen3DecoderLayer as Qwen3MoeDecoderLayer,
|
||||
)
|
||||
|
||||
# from fastdeploy.model_executor.graph_optimization.decorator import support_graph_optimization
|
||||
|
||||
|
||||
# @support_graph_optimization
|
||||
class Qwen3MoeVLModel(nn.Layer):
|
||||
"""Language backbone for Qwen3-VL."""
|
||||
|
||||
def __init__(self, fd_config: FDConfig) -> None:
|
||||
super().__init__()
|
||||
|
||||
self.num_layers = fd_config.model_config.num_hidden_layers
|
||||
self.image_token_id = fd_config.model_config.image_token_id
|
||||
self.video_token_id = fd_config.model_config.video_token_id
|
||||
self._dtype = fd_config.model_config.dtype
|
||||
fd_config.model_config.pretrained_config.prefix_name = "model"
|
||||
self.fd_config = fd_config
|
||||
|
||||
self.embed_tokens = VocabParallelEmbedding(
|
||||
fd_config=fd_config,
|
||||
num_embeddings=fd_config.model_config.vocab_size,
|
||||
embedding_dim=fd_config.model_config.hidden_size,
|
||||
params_dtype=paddle.get_default_dtype,
|
||||
prefix=f"{fd_config.model_config.pretrained_config.prefix_name}.embed_tokens",
|
||||
)
|
||||
|
||||
self.layers = nn.LayerList(
|
||||
[
|
||||
Qwen3MoeDecoderLayer(
|
||||
fd_config=fd_config,
|
||||
prefix=f"{fd_config.model_config.pretrained_config.prefix_name}.layers.{i}",
|
||||
)
|
||||
for i in range(self.num_layers)
|
||||
]
|
||||
)
|
||||
|
||||
self.norm = RMSNorm(
|
||||
fd_config,
|
||||
hidden_size=fd_config.model_config.hidden_size,
|
||||
eps=fd_config.model_config.rms_norm_eps,
|
||||
prefix=f"{fd_config.model_config.pretrained_config.prefix_name}.norm",
|
||||
)
|
||||
|
||||
def load_state_dict(self, state_dict):
|
||||
self.embed_tokens.load_state_dict(state_dict)
|
||||
self.norm.load_state_dict(state_dict)
|
||||
for i in range(self.num_layers):
|
||||
logger.info(f"Start load layer {i}")
|
||||
self.layers[i].load_state_dict(state_dict)
|
||||
|
||||
def get_input_embeddings(self, ids_remove_padding: paddle.Tensor) -> paddle.Tensor:
|
||||
return self.embed_tokens(ids_remove_padding=ids_remove_padding)
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_embeddings: paddle.Tensor,
|
||||
ids_remove_padding: paddle.Tensor,
|
||||
image_features: Optional[paddle.Tensor],
|
||||
forward_meta: ForwardMeta,
|
||||
deepstack_inputs: Optional[List[paddle.Tensor]] = None,
|
||||
) -> paddle.Tensor:
|
||||
hidden_states = input_embeddings
|
||||
residual = None
|
||||
for layer_id, layer in enumerate(self.layers):
|
||||
hidden_states, residual = layer(
|
||||
forward_meta,
|
||||
hidden_states,
|
||||
residual,
|
||||
)
|
||||
if deepstack_inputs is not None and layer_id < len(deepstack_inputs):
|
||||
hidden_states = hidden_states + deepstack_inputs[layer_id]
|
||||
hidden_states, _ = self.norm(hidden_states, residual)
|
||||
return hidden_states
|
||||
|
||||
|
||||
@ModelRegistry.register_model_class(
|
||||
architecture="Qwen3VLMoeForConditionalGeneration",
|
||||
module_name="qwen3_vl.qwen3_vl_moe",
|
||||
category=ModelCategory.MULTIMODAL,
|
||||
primary_use=ModelCategory.MULTIMODAL,
|
||||
)
|
||||
class Qwen3VLMoeForConditionalGeneration(Qwen3VLForConditionalGeneration):
|
||||
def __init__(self, fd_config: FDConfig) -> None:
|
||||
super().__init__(fd_config)
|
||||
self.config = fd_config.model_config
|
||||
self.model = Qwen3MoeVLModel(fd_config=fd_config)
|
||||
|
||||
@classmethod
|
||||
def name(cls) -> str:
|
||||
return "Qwen3VLMoeForConditionalGeneration"
|
||||
|
||||
def get_expert_mapping(
|
||||
self,
|
||||
) -> list[tuple[str, str, int, str]]:
|
||||
# (param_name, weight_name, expert_id, shard_id)
|
||||
return FusedMoE.make_expert_params_mapping(
|
||||
num_experts=self.fd_config.model_config.num_experts,
|
||||
ckpt_gate_proj_name="gate_proj",
|
||||
ckpt_down_proj_name="down_proj",
|
||||
ckpt_up_proj_name="up_proj",
|
||||
param_gate_up_proj_name="experts.up_gate_proj_",
|
||||
param_down_proj_name="experts.down_proj_",
|
||||
)
|
||||
|
||||
def load_fused_expert_weights(
|
||||
self,
|
||||
name: str,
|
||||
params_dict: dict,
|
||||
loaded_weight: paddle.Tensor,
|
||||
shard_id: str,
|
||||
num_experts: int,
|
||||
):
|
||||
param = params_dict[name]
|
||||
weight_loader = param.weight_loader
|
||||
for expert_id in range(num_experts):
|
||||
curr_expert_weight = loaded_weight[expert_id]
|
||||
weight_loader(
|
||||
param,
|
||||
curr_expert_weight,
|
||||
shard_id=shard_id,
|
||||
expert_id=expert_id,
|
||||
)
|
||||
|
||||
@paddle.no_grad()
|
||||
def load_weights(self, weights_iterator) -> None:
|
||||
"""Load model parameters from a given weights iterator."""
|
||||
|
||||
from fastdeploy.model_executor.utils import (
|
||||
default_weight_loader,
|
||||
process_weights_after_loading,
|
||||
)
|
||||
|
||||
stacked_params_mapping = [
|
||||
# (param_name, weight_name, expert_id, shard_id)
|
||||
("qkv_proj", "q_proj", "q"),
|
||||
("qkv_proj", "k_proj", "k"),
|
||||
("qkv_proj", "v_proj", "v"),
|
||||
("up_gate_proj", "gate_proj", "gate"),
|
||||
("up_gate_proj", "up_proj", "up"),
|
||||
("embed_tokens.embeddings", "embed_tokens", None),
|
||||
("lm_head.linear", "lm_head", None),
|
||||
("visual", "model.visual", None),
|
||||
]
|
||||
|
||||
expert_params_mapping = self.get_expert_mapping() # Not actually used
|
||||
params_dict = dict(self.named_parameters())
|
||||
is_fused_expert = False
|
||||
fused_expert_params_mapping = [
|
||||
("experts.up_gate_proj_weight", "experts.gate_up_proj", 0, "gate"),
|
||||
("experts.down_proj_weight", "experts.down_proj", 0, "down"),
|
||||
]
|
||||
num_experts = self.config.num_experts
|
||||
# params_name model.embed_tokens.embeddings.weight
|
||||
# weight_name model.language_model.embed_tokens.weight
|
||||
process_weights_after_loading_fn = process_weights_after_loading(dict(self.named_sublayers()), self.fd_config)
|
||||
logger.info(f"[Qwen3Moe-VL] params_dict names: {list(params_dict.keys())} ")
|
||||
for loaded_weight_name, loaded_weight in weights_iterator:
|
||||
loaded_weight_name = loaded_weight_name.replace(".language_model", "")
|
||||
for param_name, weight_name, shard_id in stacked_params_mapping:
|
||||
if "experts.gate_up_proj" in loaded_weight_name or "experts.down_proj" in loaded_weight_name:
|
||||
is_fused_expert = True
|
||||
expert_params_mapping = fused_expert_params_mapping
|
||||
|
||||
if weight_name not in loaded_weight_name:
|
||||
continue
|
||||
if "mlp.experts" in loaded_weight_name:
|
||||
continue
|
||||
model_param_name = loaded_weight_name.replace(weight_name, param_name)
|
||||
if model_param_name not in params_dict:
|
||||
continue
|
||||
param = params_dict[model_param_name]
|
||||
weight_loader = getattr(param, "weight_loader", default_weight_loader(self.fd_config))
|
||||
weight_loader(param, loaded_weight, shard_id)
|
||||
break
|
||||
else:
|
||||
is_expert_weight = False
|
||||
# Note: mlp.experts.gate_up_proj in qwen3moe_vl is merged and should be processed separately when loading weights
|
||||
for mapping in expert_params_mapping:
|
||||
param_name, weight_name, expert_id, shard_id = mapping
|
||||
if weight_name not in loaded_weight_name:
|
||||
continue
|
||||
# Anyway, this is an expert weight and should not be
|
||||
# attempted to load as other weights later
|
||||
is_expert_weight = True
|
||||
model_param_name = loaded_weight_name.replace(weight_name, param_name)
|
||||
if is_fused_expert:
|
||||
loaded_weight = loaded_weight.transpose(-1, -2)
|
||||
if "experts.gate_up_proj" in loaded_weight_name:
|
||||
gate_weight, up_weight = loaded_weight.chunk(2, dim=-2)
|
||||
self.load_fused_expert_weights(
|
||||
model_param_name,
|
||||
params_dict,
|
||||
gate_weight,
|
||||
"gate",
|
||||
num_experts,
|
||||
)
|
||||
self.load_fused_expert_weights(
|
||||
model_param_name,
|
||||
params_dict,
|
||||
up_weight,
|
||||
"up",
|
||||
num_experts,
|
||||
)
|
||||
else:
|
||||
# down_proj
|
||||
self.load_fused_expert_weights(
|
||||
model_param_name,
|
||||
params_dict,
|
||||
loaded_weight,
|
||||
shard_id,
|
||||
num_experts,
|
||||
)
|
||||
break
|
||||
else:
|
||||
|
||||
if weight_name not in loaded_weight_name:
|
||||
continue
|
||||
model_param_name = loaded_weight_name.replace(weight_name, param_name)
|
||||
if model_param_name not in params_dict:
|
||||
continue
|
||||
param = params_dict[model_param_name]
|
||||
weight_loader = param.weight_loader
|
||||
weight_loader(param, loaded_weight, shard_id=shard_id, expert_id=expert_id)
|
||||
break
|
||||
else:
|
||||
if is_expert_weight:
|
||||
continue
|
||||
model_param_name = loaded_weight_name
|
||||
if model_param_name not in params_dict:
|
||||
continue
|
||||
param = params_dict[model_param_name]
|
||||
weight_loader = getattr(param, "weight_loader", default_weight_loader(self.fd_config))
|
||||
weight_loader(param, loaded_weight)
|
||||
|
||||
model_sublayer_name = re.sub(r"\.(up_gate_proj_weight|down_proj_weight|weight)$", "", model_param_name)
|
||||
process_weights_after_loading_fn(model_sublayer_name, param)
|
||||
|
||||
if self.tie_word_embeddings:
|
||||
self.lm_head.linear.weight.set_value(
|
||||
self.model.embed_tokens.embeddings.weight.transpose([1, 0]).astype(self.lm_head.linear.weight.dtype)
|
||||
)
|
||||
Reference in New Issue
Block a user