SUMIN/HuggingFace_transformer
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity

Adding Qwen3 and Qwen3MoE (#36878)

Browse Source 
* Initial commit for Qwen3

* fix and add tests for qwen3 & qwen3_moe

* rename models for tests.

* fix

* fix

* fix and add docs.

* fix model name in docs.

* simplify modular and fix configuration issues

* Fix the red CI: ruff was updated

* revert ruff, version was wrong

* fix qwen3moe.

* fix

* make sure MOE can load

* fix copies

---------

Co-authored-by: Arthur Zucker <arthur.zucker@gmail.com>
This commit is contained in:
Bo Zheng
2025-03-31 15:50:49 +08:00
committed by GitHub
parent 0d6a60fe55
commit 6acd5aecb3
26 changed files with 5650 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

40
src/transformers/__init__.py
View File

@@ -744,6 +744,8 @@ _import_structure = {
"Qwen2VLConfig",
"Qwen2VLProcessor",
],
"models.qwen3": ["Qwen3Config"],
"models.qwen3_moe": ["Qwen3MoeConfig"],
"models.rag": ["RagConfig", "RagRetriever", "RagTokenizer"],
"models.recurrent_gemma": ["RecurrentGemmaConfig"],
"models.reformer": ["ReformerConfig"],
@@ -3441,6 +3443,26 @@ else:
"Qwen2VLPreTrainedModel",
]
)
_import_structure["models.qwen3"].extend(
[
"Qwen3ForCausalLM",
"Qwen3ForQuestionAnswering",
"Qwen3ForSequenceClassification",
"Qwen3ForTokenClassification",
"Qwen3Model",
"Qwen3PreTrainedModel",
]
)
_import_structure["models.qwen3_moe"].extend(
[
"Qwen3MoeForCausalLM",
"Qwen3MoeForQuestionAnswering",
"Qwen3MoeForSequenceClassification",
"Qwen3MoeForTokenClassification",
"Qwen3MoeModel",
"Qwen3MoePreTrainedModel",
]
)
_import_structure["models.rag"].extend(
[
"RagModel",
@@ -5993,6 +6015,8 @@ if TYPE_CHECKING:
Qwen2VLConfig,
Qwen2VLProcessor,
)
from .models.qwen3 import Qwen3Config
from .models.qwen3_moe import Qwen3MoeConfig
from .models.rag import RagConfig, RagRetriever, RagTokenizer
from .models.recurrent_gemma import RecurrentGemmaConfig
from .models.reformer import ReformerConfig
@@ -8293,6 +8317,22 @@ if TYPE_CHECKING:
Qwen2VLModel,
Qwen2VLPreTrainedModel,
)
from .models.qwen3 import (
Qwen3ForCausalLM,
Qwen3ForQuestionAnswering,
Qwen3ForSequenceClassification,
Qwen3ForTokenClassification,
Qwen3Model,
Qwen3PreTrainedModel,
)
from .models.qwen3_moe import (
Qwen3MoeForCausalLM,
Qwen3MoeForQuestionAnswering,
Qwen3MoeForSequenceClassification,
Qwen3MoeForTokenClassification,
Qwen3MoeModel,
Qwen3MoePreTrainedModel,
)
from .models.rag import (
RagModel,
RagPreTrainedModel,

2
src/transformers/models/__init__.py
View File

@@ -230,6 +230,8 @@ from . import (
qwen2_audio,
qwen2_moe,
qwen2_vl,
qwen3,
qwen3_moe,
rag,
recurrent_gemma,
reformer,

4
src/transformers/models/auto/configuration_auto.py
View File

@@ -254,6 +254,8 @@ CONFIG_MAPPING_NAMES = OrderedDict(
("qwen2_audio_encoder", "Qwen2AudioEncoderConfig"),
("qwen2_moe", "Qwen2MoeConfig"),
("qwen2_vl", "Qwen2VLConfig"),
("qwen3", "Qwen3Config"),
("qwen3_moe", "Qwen3MoeConfig"),
("rag", "RagConfig"),
("realm", "RealmConfig"),
("recurrent_gemma", "RecurrentGemmaConfig"),
@@ -609,6 +611,8 @@ MODEL_NAMES_MAPPING = OrderedDict(
("qwen2_audio_encoder", "Qwen2AudioEncoder"),
("qwen2_moe", "Qwen2MoE"),
("qwen2_vl", "Qwen2VL"),
("qwen3", "Qwen3"),
("qwen3_moe", "Qwen3MoE"),
("rag", "RAG"),
("realm", "REALM"),
("recurrent_gemma", "RecurrentGemma"),

10
src/transformers/models/auto/modeling_auto.py
View File

@@ -233,6 +233,8 @@ MODEL_MAPPING_NAMES = OrderedDict(
("qwen2_audio_encoder", "Qwen2AudioEncoder"),
("qwen2_moe", "Qwen2MoeModel"),
("qwen2_vl", "Qwen2VLModel"),
("qwen3", "Qwen3Model"),
("qwen3_moe", "Qwen3MoeModel"),
("recurrent_gemma", "RecurrentGemmaModel"),
("reformer", "ReformerModel"),
("regnet", "RegNetModel"),
@@ -576,6 +578,8 @@ MODEL_FOR_CAUSAL_LM_MAPPING_NAMES = OrderedDict(
("qdqbert", "QDQBertLMHeadModel"),
("qwen2", "Qwen2ForCausalLM"),
("qwen2_moe", "Qwen2MoeForCausalLM"),
("qwen3", "Qwen3ForCausalLM"),
("qwen3_moe", "Qwen3MoeForCausalLM"),
("recurrent_gemma", "RecurrentGemmaForCausalLM"),
("reformer", "ReformerModelWithLMHead"),
("rembert", "RemBertForCausalLM"),
@@ -1072,6 +1076,8 @@ MODEL_FOR_SEQUENCE_CLASSIFICATION_MAPPING_NAMES = OrderedDict(
("qdqbert", "QDQBertForSequenceClassification"),
("qwen2", "Qwen2ForSequenceClassification"),
("qwen2_moe", "Qwen2MoeForSequenceClassification"),
("qwen3", "Qwen3ForSequenceClassification"),
("qwen3_moe", "Qwen3MoeForSequenceClassification"),
("reformer", "ReformerForSequenceClassification"),
("rembert", "RemBertForSequenceClassification"),
("roberta", "RobertaForSequenceClassification"),
@@ -1153,6 +1159,8 @@ MODEL_FOR_QUESTION_ANSWERING_MAPPING_NAMES = OrderedDict(
("qdqbert", "QDQBertForQuestionAnswering"),
("qwen2", "Qwen2ForQuestionAnswering"),
("qwen2_moe", "Qwen2MoeForQuestionAnswering"),
("qwen3", "Qwen3ForQuestionAnswering"),
("qwen3_moe", "Qwen3MoeForQuestionAnswering"),
("reformer", "ReformerForQuestionAnswering"),
("rembert", "RemBertForQuestionAnswering"),
("roberta", "RobertaForQuestionAnswering"),
@@ -1257,6 +1265,8 @@ MODEL_FOR_TOKEN_CLASSIFICATION_MAPPING_NAMES = OrderedDict(
("qdqbert", "QDQBertForTokenClassification"),
("qwen2", "Qwen2ForTokenClassification"),
("qwen2_moe", "Qwen2MoeForTokenClassification"),
("qwen3", "Qwen3ForTokenClassification"),
("qwen3_moe", "Qwen3MoeForTokenClassification"),
("rembert", "RemBertForTokenClassification"),
("roberta", "RobertaForTokenClassification"),
("roberta-prelayernorm", "RobertaPreLayerNormForTokenClassification"),

14
src/transformers/models/auto/tokenization_auto.py
View File

@@ -454,6 +454,20 @@ else:
),
),
("qwen2_vl", ("Qwen2Tokenizer", "Qwen2TokenizerFast" if is_tokenizers_available() else None)),
(
"qwen3",
(
"Qwen2Tokenizer",
"Qwen2TokenizerFast" if is_tokenizers_available() else None,
),
),
(
"qwen3_moe",
(
"Qwen2Tokenizer",
"Qwen2TokenizerFast" if is_tokenizers_available() else None,
),
),
("rag", ("RagTokenizer", None)),
("realm", ("RealmTokenizer", "RealmTokenizerFast" if is_tokenizers_available() else None)),
(

3
src/transformers/models/qwen2_moe/configuration_qwen2_moe.py
View File

@@ -26,8 +26,7 @@ class Qwen2MoeConfig(PretrainedConfig):
r"""
This is the configuration class to store the configuration of a [`Qwen2MoeModel`]. It is used to instantiate a
Qwen2MoE model according to the specified arguments, defining the model architecture. Instantiating a configuration
with the defaults will yield a similar configuration to that of
Qwen1.5-MoE-A2.7B" [Qwen/Qwen1.5-MoE-A2.7B"](https://huggingface.co/Qwen/Qwen1.5-MoE-A2.7B").
with the defaults will yield a similar configuration to that of [Qwen/Qwen1.5-MoE-A2.7B](https://huggingface.co/Qwen/Qwen1.5-MoE-A2.7B).
Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
documentation from [`PretrainedConfig`] for more information.

27
src/transformers/models/qwen3/__init__.py Normal file
View File

@@ -0,0 +1,27 @@
# Copyright 2024 The Qwen Team and The HuggingFace Inc. team. 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 typing import TYPE_CHECKING
from ...utils import _LazyModule
from ...utils.import_utils import define_import_structure
if TYPE_CHECKING:
from .configuration_qwen3 import *
from .modeling_qwen3 import *
else:
import sys
_file = globals()["__file__"]
sys.modules[__name__] = _LazyModule(__name__, _file, define_import_structure(_file), module_spec=__spec__)

212
src/transformers/models/qwen3/configuration_qwen3.py Normal file
View File

@@ -0,0 +1,212 @@
# coding=utf-8
# Copyright 2024 The Qwen team, Alibaba Group and the HuggingFace Inc. team. 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.
"""Qwen3 model configuration"""
from ...configuration_utils import PretrainedConfig
from ...modeling_rope_utils import rope_config_validation
from ...utils import logging
logger = logging.get_logger(__name__)
class Qwen3Config(PretrainedConfig):
r"""
This is the configuration class to store the configuration of a [`Qwen3Model`]. It is used to instantiate a
Qwen3 model according to the specified arguments, defining the model architecture. Instantiating a configuration
with the defaults will yield a similar configuration to that of
Qwen3-8B [Qwen/Qwen3-8B](https://huggingface.co/Qwen/Qwen3-8B).
Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
documentation from [`PretrainedConfig`] for more information.
Args:
vocab_size (`int`, *optional*, defaults to 151936):
Vocabulary size of the Qwen3 model. Defines the number of different tokens that can be represented by the
`inputs_ids` passed when calling [`Qwen3Model`]
hidden_size (`int`, *optional*, defaults to 4096):
Dimension of the hidden representations.
intermediate_size (`int`, *optional*, defaults to 22016):
Dimension of the MLP representations.
num_hidden_layers (`int`, *optional*, defaults to 32):
Number of hidden layers in the Transformer encoder.
num_attention_heads (`int`, *optional*, defaults to 32):
Number of attention heads for each attention layer in the Transformer encoder.
num_key_value_heads (`int`, *optional*, defaults to 32):
This is the number of key_value heads that should be used to implement Grouped Query Attention. If
`num_key_value_heads=num_attention_heads`, the model will use Multi Head Attention (MHA), if
`num_key_value_heads=1` the model will use Multi Query Attention (MQA) otherwise GQA is used. When
converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed
by meanpooling all the original heads within that group. For more details checkout [this
paper](https://arxiv.org/pdf/2305.13245.pdf). If it is not specified, will default to `32`.
head_dim (`int`, *optional*, defaults to 128):
The attention head dimension.
hidden_act (`str` or `function`, *optional*, defaults to `"silu"`):
The non-linear activation function (function or string) in the decoder.
max_position_embeddings (`int`, *optional*, defaults to 32768):
The maximum sequence length that this model might ever be used with.
initializer_range (`float`, *optional*, defaults to 0.02):
The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
rms_norm_eps (`float`, *optional*, defaults to 1e-06):
The epsilon used by the rms normalization layers.
use_cache (`bool`, *optional*, defaults to `True`):
Whether or not the model should return the last key/values attentions (not used by all models). Only
relevant if `config.is_decoder=True`.
tie_word_embeddings (`bool`, *optional*, defaults to `False`):
Whether the model's input and output word embeddings should be tied.
rope_theta (`float`, *optional*, defaults to 10000.0):
The base period of the RoPE embeddings.
rope_scaling (`Dict`, *optional*):
Dictionary containing the scaling configuration for the RoPE embeddings. NOTE: if you apply new rope type
and you expect the model to work on longer `max_position_embeddings`, we recommend you to update this value
accordingly.
Expected contents:
`rope_type` (`str`):
The sub-variant of RoPE to use. Can be one of ['default', 'linear', 'dynamic', 'yarn', 'longrope',
'llama3'], with 'default' being the original RoPE implementation.
`factor` (`float`, *optional*):
Used with all rope types except 'default'. The scaling factor to apply to the RoPE embeddings. In
most scaling types, a `factor` of x will enable the model to handle sequences of length x *
original maximum pre-trained length.
`original_max_position_embeddings` (`int`, *optional*):
Used with 'dynamic', 'longrope' and 'llama3'. The original max position embeddings used during
pretraining.
`attention_factor` (`float`, *optional*):
Used with 'yarn' and 'longrope'. The scaling factor to be applied on the attention
computation. If unspecified, it defaults to value recommended by the implementation, using the
`factor` field to infer the suggested value.
`beta_fast` (`float`, *optional*):
Only used with 'yarn'. Parameter to set the boundary for extrapolation (only) in the linear
ramp function. If unspecified, it defaults to 32.
`beta_slow` (`float`, *optional*):
Only used with 'yarn'. Parameter to set the boundary for interpolation (only) in the linear
ramp function. If unspecified, it defaults to 1.
`short_factor` (`List[float]`, *optional*):
Only used with 'longrope'. The scaling factor to be applied to short contexts (<
`original_max_position_embeddings`). Must be a list of numbers with the same length as the hidden
size divided by the number of attention heads divided by 2
`long_factor` (`List[float]`, *optional*):
Only used with 'longrope'. The scaling factor to be applied to long contexts (<
`original_max_position_embeddings`). Must be a list of numbers with the same length as the hidden
size divided by the number of attention heads divided by 2
`low_freq_factor` (`float`, *optional*):
Only used with 'llama3'. Scaling factor applied to low frequency components of the RoPE
`high_freq_factor` (`float`, *optional*):
Only used with 'llama3'. Scaling factor applied to high frequency components of the RoPE
attention_bias (`bool`, defaults to `False`, *optional*, defaults to `False`):
Whether to use a bias in the query, key, value and output projection layers during self-attention.
use_sliding_window (`bool`, *optional*, defaults to `False`):
Whether to use sliding window attention.
sliding_window (`int`, *optional*, defaults to 4096):
Sliding window attention (SWA) window size. If not specified, will default to `4096`.
max_window_layers (`int`, *optional*, defaults to 28):
The number of layers that use SWA (Sliding Window Attention). The bottom layers use SWA while the top use full attention.
attention_dropout (`float`, *optional*, defaults to 0.0):
The dropout ratio for the attention probabilities.
```python
>>> from transformers import Qwen3Model, Qwen3Config
>>> # Initializing a Qwen3 style configuration
>>> configuration = Qwen3Config()
>>> # Initializing a model from the Qwen3-8B style configuration
>>> model = Qwen3Model(configuration)
>>> # Accessing the model configuration
>>> configuration = model.config
```"""
model_type = "qwen3"
keys_to_ignore_at_inference = ["past_key_values"]
# Default tensor parallel plan for base model `Qwen3`
base_model_tp_plan = {
"layers.*.self_attn.q_proj": "colwise",
"layers.*.self_attn.k_proj": "colwise",
"layers.*.self_attn.v_proj": "colwise",
"layers.*.self_attn.o_proj": "rowwise",
"layers.*.mlp.gate_proj": "colwise",
"layers.*.mlp.up_proj": "colwise",
"layers.*.mlp.down_proj": "rowwise",
}
base_model_pp_plan = {
"embed_tokens": (["input_ids"], ["inputs_embeds"]),
"layers": (["hidden_states", "attention_mask"], ["hidden_states"]),
"norm": (["hidden_states"], ["hidden_states"]),
}
def __init__(
self,
vocab_size=151936,
hidden_size=4096,
intermediate_size=22016,
num_hidden_layers=32,
num_attention_heads=32,
num_key_value_heads=32,
head_dim=128,
hidden_act="silu",
max_position_embeddings=32768,
initializer_range=0.02,
rms_norm_eps=1e-6,
use_cache=True,
tie_word_embeddings=False,
rope_theta=10000.0,
rope_scaling=None,
attention_bias=False,
use_sliding_window=False,
sliding_window=4096,
max_window_layers=28,
attention_dropout=0.0,
**kwargs,
):
self.vocab_size = vocab_size
self.max_position_embeddings = max_position_embeddings
self.hidden_size = hidden_size
self.intermediate_size = intermediate_size
self.num_hidden_layers = num_hidden_layers
self.num_attention_heads = num_attention_heads
self.use_sliding_window = use_sliding_window
self.sliding_window = sliding_window # we check `use_sliding_window` in the modeling code
self.max_window_layers = max_window_layers
# for backward compatibility
if num_key_value_heads is None:
num_key_value_heads = num_attention_heads
self.num_key_value_heads = num_key_value_heads
self.head_dim = head_dim
self.hidden_act = hidden_act
self.initializer_range = initializer_range
self.rms_norm_eps = rms_norm_eps
self.use_cache = use_cache
self.rope_theta = rope_theta
self.rope_scaling = rope_scaling
self.attention_bias = attention_bias
self.attention_dropout = attention_dropout
# Validate the correctness of rotary position embeddings parameters
# BC: if there is a 'type' field, move it to 'rope_type'.
if self.rope_scaling is not None and "type" in self.rope_scaling:
self.rope_scaling["rope_type"] = self.rope_scaling["type"]
rope_config_validation(self)
super().__init__(
tie_word_embeddings=tie_word_embeddings,
**kwargs,
)
__all__ = ["Qwen3Config"]

1202
src/transformers/models/qwen3/modeling_qwen3.py Normal file
View File

File diff suppressed because it is too large Load Diff

203
src/transformers/models/qwen3/modular_qwen3.py Normal file
View File

@@ -0,0 +1,203 @@
# coding=utf-8
# Copyright 2025 The Qwen team, Alibaba Group and the HuggingFace Inc. team. 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.
"""PyTorch Qwen3 model."""
from typing import Callable, Optional, Tuple, Union
import torch
import torch.utils.checkpoint
from ...cache_utils import Cache
from ...modeling_flash_attention_utils import FlashAttentionKwargs
from ...modeling_outputs import CausalLMOutputWithPast
from ...modeling_utils import ALL_ATTENTION_FUNCTIONS
from ...processing_utils import Unpack
from ...utils import (
LossKwargs,
logging,
)
from ..gemma.modeling_gemma import GemmaMLP
from ..llama.modeling_llama import (
LlamaAttention,
LlamaDecoderLayer,
LlamaForCausalLM,
LlamaForQuestionAnswering,
LlamaForSequenceClassification,
LlamaForTokenClassification,
LlamaRMSNorm,
apply_rotary_pos_emb,
eager_attention_forward,
)
from ..mistral.modeling_mistral import MistralModel
from .configuration_qwen3 import Qwen3Config
logger = logging.get_logger(__name__)
_CHECKPOINT_FOR_DOC = "Qwen/Qwen3-8B"
class Qwen3RMSNorm(LlamaRMSNorm):
pass
class Qwen3MLP(GemmaMLP):
pass
class Qwen3Attention(LlamaAttention):
def __init__(self, config: Qwen3Config, layer_idx: int):
super().__init__(config, layer_idx)
self.q_norm = Qwen3RMSNorm(self.head_dim, eps=config.rms_norm_eps) # unlike olmo, only on the head dim!
self.k_norm = Qwen3RMSNorm(self.head_dim, eps=config.rms_norm_eps) # thus post q_norm does not need reshape
self.sliding_window = config.sliding_window
if not (
self.config.use_sliding_window
and getattr(self.config, "sliding_window", None) is not None
and self.layer_idx >= self.config.max_window_layers
):
self.sliding_window = None
def forward(
self,
hidden_states: torch.Tensor,
position_embeddings: Tuple[torch.Tensor, torch.Tensor],
attention_mask: Optional[torch.Tensor],
past_key_value: Optional[Cache] = None,
cache_position: Optional[torch.LongTensor] = None,
**kwargs: Unpack[FlashAttentionKwargs],
) -> Tuple[torch.Tensor, Optional[torch.Tensor], Optional[Tuple[torch.Tensor]]]:
input_shape = hidden_states.shape[:-1]
hidden_shape = (*input_shape, -1, self.head_dim)
query_states = self.q_norm(self.q_proj(hidden_states).view(hidden_shape)).transpose(1, 2)
key_states = self.k_norm(self.k_proj(hidden_states).view(hidden_shape)).transpose(1, 2)
value_states = self.v_proj(hidden_states).view(hidden_shape).transpose(1, 2)
cos, sin = position_embeddings
query_states, key_states = apply_rotary_pos_emb(query_states, key_states, cos, sin)
if past_key_value is not None:
# sin and cos are specific to RoPE models; cache_position needed for the static cache
cache_kwargs = {"sin": sin, "cos": cos, "cache_position": cache_position}
key_states, value_states = past_key_value.update(key_states, value_states, self.layer_idx, cache_kwargs)
attention_interface: Callable = eager_attention_forward
if self.config._attn_implementation != "eager":
if self.config._attn_implementation == "sdpa" and kwargs.get("output_attentions", False):
logger.warning_once(
"`torch.nn.functional.scaled_dot_product_attention` does not support `output_attentions=True`. Falling back to "
'eager attention. This warning can be removed using the argument `attn_implementation="eager"` when loading the model.'
)
else:
attention_interface = ALL_ATTENTION_FUNCTIONS[self.config._attn_implementation]
attn_output, attn_weights = attention_interface(
self,
query_states,
key_states,
value_states,
attention_mask,
dropout=0.0 if not self.training else self.attention_dropout,
scaling=self.scaling,
sliding_window=self.sliding_window, # diff with Llama
**kwargs,
)
attn_output = attn_output.reshape(*input_shape, -1).contiguous()
attn_output = self.o_proj(attn_output)
return attn_output, attn_weights
class Qwen3DecoderLayer(LlamaDecoderLayer):
def __init__(self, config: Qwen3Config, layer_idx: int):
super().__init__()
self.self_attn = Qwen3Attention(config=config, layer_idx=layer_idx)
self.mlp = Qwen3MLP(config)
if (
config.sliding_window and config._attn_implementation != "flash_attention_2"
): # diff with Llama is this warning
logger.warning_once(
f"Sliding Window Attention is enabled but not implemented for `{config._attn_implementation}`; "
"unexpected results may be encountered."
)
class Qwen3Model(MistralModel): # mistral model creates sliding window
pass
class KwargsForCausalLM(FlashAttentionKwargs, LossKwargs): ...
class Qwen3ForCausalLM(LlamaForCausalLM):
def forward(
self,
**super_kwargs: Unpack[KwargsForCausalLM],
) -> Union[Tuple, CausalLMOutputWithPast]:
r"""
labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
Labels for computing the masked language modeling loss. Indices should either be in `[0, ...,
config.vocab_size]` or -100 (see `input_ids` docstring). Tokens with indices set to `-100` are ignored
(masked), the loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`.
logits_to_keep (`int` or `torch.Tensor`, *optional*):
If an `int`, compute logits for the last `logits_to_keep` tokens. If `0`, calculate logits for all
`input_ids` (special case). Only last token logits are needed for generation, and calculating them only for that
token can save memory, which becomes pretty significant for long sequences or large vocabulary size.
If a `torch.Tensor`, must be 1D corresponding to the indices to keep in the sequence length dimension.
This is useful when using packed tensor format (single dimension for batch and sequence length).
Returns:
Example:
```python
>>> from transformers import AutoTokenizer, Qwen3ForCausalLM
>>> model = Qwen3ForCausalLM.from_pretrained("Qwen/Qwen3-8B")
>>> tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen3-8B")
>>> prompt = "Hey, are you conscious? Can you talk to me?"
>>> inputs = tokenizer(prompt, return_tensors="pt")
>>> # Generate
>>> generate_ids = model.generate(inputs.input_ids, max_length=30)
>>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
"Hey, are you conscious? Can you talk to me?\nI'm not conscious, but I can talk to you."
```"""
return super().forward(**super_kwargs)
class Qwen3ForSequenceClassification(LlamaForSequenceClassification):
pass
class Qwen3ForTokenClassification(LlamaForTokenClassification):
pass
class Qwen3ForQuestionAnswering(LlamaForQuestionAnswering):
pass
__all__ = [
"Qwen3ForCausalLM",
"Qwen3ForQuestionAnswering",
"Qwen3Model",
"Qwen3PreTrainedModel", # noqa: F822
"Qwen3ForSequenceClassification",
"Qwen3ForTokenClassification",
]

27
src/transformers/models/qwen3_moe/__init__.py Normal file
View File

@@ -0,0 +1,27 @@
# Copyright 2024 The HuggingFace Team. 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 typing import TYPE_CHECKING
from ...utils import _LazyModule
from ...utils.import_utils import define_import_structure
if TYPE_CHECKING:
from .configuration_qwen3_moe import *
from .modeling_qwen3_moe import *
else:
import sys
_file = globals()["__file__"]
sys.modules[__name__] = _LazyModule(__name__, _file, define_import_structure(_file), module_spec=__spec__)

240
src/transformers/models/qwen3_moe/configuration_qwen3_moe.py Normal file
View File

@@ -0,0 +1,240 @@
# coding=utf-8
# Copyright 2024 The Qwen team, Alibaba Group and the HuggingFace Inc. team. 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.
"""Qwen3MoE model configuration"""
from ...configuration_utils import PretrainedConfig
from ...modeling_rope_utils import rope_config_validation
from ...utils import logging
logger = logging.get_logger(__name__)
class Qwen3MoeConfig(PretrainedConfig):
r"""
This is the configuration class to store the configuration of a [`Qwen3MoeModel`]. It is used to instantiate a
Qwen3MoE model according to the specified arguments, defining the model architecture. Instantiating a configuration
with the defaults will yield a similar configuration to that of [Qwen/Qwen3-MoE-15B-A2B](https://huggingface.co/Qwen/Qwen3-15B-A2B).
Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
documentation from [`PretrainedConfig`] for more information.
Args:
vocab_size (`int`, *optional*, defaults to 151936):
Vocabulary size of the Qwen3MoE model. Defines the number of different tokens that can be represented by the
`inputs_ids` passed when calling [`Qwen3MoeModel`]
hidden_size (`int`, *optional*, defaults to 2048):
Dimension of the hidden representations.
intermediate_size (`int`, *optional*, defaults to 6144):
Dimension of the MLP representations.
num_hidden_layers (`int`, *optional*, defaults to 24):
Number of hidden layers in the Transformer encoder.
num_attention_heads (`int`, *optional*, defaults to 32):
Number of attention heads for each attention layer in the Transformer encoder.
num_key_value_heads (`int`, *optional*, defaults to 4):
This is the number of key_value heads that should be used to implement Grouped Query Attention. If
`num_key_value_heads=num_attention_heads`, the model will use Multi Head Attention (MHA), if
`num_key_value_heads=1` the model will use Multi Query Attention (MQA) otherwise GQA is used. When
converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed
by meanpooling all the original heads within that group. For more details checkout [this
paper](https://arxiv.org/pdf/2305.13245.pdf). If it is not specified, will default to `32`.
hidden_act (`str` or `function`, *optional*, defaults to `"silu"`):
The non-linear activation function (function or string) in the decoder.
max_position_embeddings (`int`, *optional*, defaults to 32768):
The maximum sequence length that this model might ever be used with.
initializer_range (`float`, *optional*, defaults to 0.02):
The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
rms_norm_eps (`float`, *optional*, defaults to 1e-06):
The epsilon used by the rms normalization layers.
use_cache (`bool`, *optional*, defaults to `True`):
Whether or not the model should return the last key/values attentions (not used by all models). Only
relevant if `config.is_decoder=True`.
tie_word_embeddings (`bool`, *optional*, defaults to `False`):
Whether the model's input and output word embeddings should be tied.
rope_theta (`float`, *optional*, defaults to 10000.0):
The base period of the RoPE embeddings.
rope_scaling (`Dict`, *optional*):
Dictionary containing the scaling configuration for the RoPE embeddings. NOTE: if you apply new rope type
and you expect the model to work on longer `max_position_embeddings`, we recommend you to update this value
accordingly.
Expected contents:
`rope_type` (`str`):
The sub-variant of RoPE to use. Can be one of ['default', 'linear', 'dynamic', 'yarn', 'longrope',
'llama3'], with 'default' being the original RoPE implementation.
`factor` (`float`, *optional*):
Used with all rope types except 'default'. The scaling factor to apply to the RoPE embeddings. In
most scaling types, a `factor` of x will enable the model to handle sequences of length x *
original maximum pre-trained length.
`original_max_position_embeddings` (`int`, *optional*):
Used with 'dynamic', 'longrope' and 'llama3'. The original max position embeddings used during
pretraining.
`attention_factor` (`float`, *optional*):
Used with 'yarn' and 'longrope'. The scaling factor to be applied on the attention
computation. If unspecified, it defaults to value recommended by the implementation, using the
`factor` field to infer the suggested value.
`beta_fast` (`float`, *optional*):
Only used with 'yarn'. Parameter to set the boundary for extrapolation (only) in the linear
ramp function. If unspecified, it defaults to 32.
`beta_slow` (`float`, *optional*):
Only used with 'yarn'. Parameter to set the boundary for interpolation (only) in the linear
ramp function. If unspecified, it defaults to 1.
`short_factor` (`List[float]`, *optional*):
Only used with 'longrope'. The scaling factor to be applied to short contexts (<
`original_max_position_embeddings`). Must be a list of numbers with the same length as the hidden
size divided by the number of attention heads divided by 2
`long_factor` (`List[float]`, *optional*):
Only used with 'longrope'. The scaling factor to be applied to long contexts (<
`original_max_position_embeddings`). Must be a list of numbers with the same length as the hidden
size divided by the number of attention heads divided by 2
`low_freq_factor` (`float`, *optional*):
Only used with 'llama3'. Scaling factor applied to low frequency components of the RoPE
`high_freq_factor` (`float`, *optional*):
Only used with 'llama3'. Scaling factor applied to high frequency components of the RoPE
attention_bias (`bool`, defaults to `False`, *optional*, defaults to `False`):
Whether to use a bias in the query, key, value and output projection layers during self-attention.
use_sliding_window (`bool`, *optional*, defaults to `False`):
Whether to use sliding window attention.
sliding_window (`int`, *optional*, defaults to 4096):
Sliding window attention (SWA) window size. If not specified, will default to `4096`.
max_window_layers (`int`, *optional*, defaults to 28):
The number of layers that use SWA (Sliding Window Attention). The bottom layers use SWA while the top use full attention.
attention_dropout (`float`, *optional*, defaults to 0.0):
The dropout ratio for the attention probabilities.
decoder_sparse_step (`int`, *optional*, defaults to 1):
The frequency of the MoE layer.
moe_intermediate_size (`int`, *optional*, defaults to 768):
Intermediate size of the routed expert.
num_experts_per_tok (`int`, *optional*, defaults to 8):
Number of selected experts.
num_experts (`int`, *optional*, defaults to 128):
Number of routed experts.
norm_topk_prob (`bool`, *optional*, defaults to `False`):
Whether to normalize the topk probabilities.
output_router_logits (`bool`, *optional*, defaults to `False`):
Whether or not the router logits should be returned by the model. Enabeling this will also
allow the model to output the auxiliary loss, including load balancing loss and router z-loss.
router_aux_loss_coef (`float`, *optional*, defaults to 0.001):
The aux loss factor for the total loss.
mlp_only_layers (`List[int]`, *optional*, defaults to `[]`):
Indicate which layers use Qwen3MoeMLP rather than Qwen3MoeSparseMoeBlock
The list contains layer index, from 0 to num_layers-1 if we have num_layers layers
If `mlp_only_layers` is empty, `decoder_sparse_step` is used to determine the sparsity.
```python
>>> from transformers import Qwen3MoeModel, Qwen3MoeConfig
>>> # Initializing a Qwen3MoE style configuration
>>> configuration = Qwen3MoeConfig()
>>> # Initializing a model from the Qwen3-15B-A2B" style configuration
>>> model = Qwen3MoeModel(configuration)
>>> # Accessing the model configuration
>>> configuration = model.config
```"""
model_type = "qwen3_moe"
keys_to_ignore_at_inference = ["past_key_values"]
# Default tensor parallel plan for base model `Qwen3Moe`
base_model_tp_plan = {
"layers.*.self_attn.q_proj": "colwise",
"layers.*.self_attn.k_proj": "colwise",
"layers.*.self_attn.v_proj": "colwise",
"layers.*.self_attn.o_proj": "rowwise",
"layers.*.mlp.gate_proj": "colwise",
"layers.*.mlp.up_proj": "colwise",
"layers.*.mlp.down_proj": "rowwise",
}
base_model_pp_plan = {
"embed_tokens": (["input_ids"], ["inputs_embeds"]),
"layers": (["hidden_states", "attention_mask"], ["hidden_states"]),
"norm": (["hidden_states"], ["hidden_states"]),
}
def __init__(
self,
vocab_size=151936,
hidden_size=2048,
intermediate_size=6144,
num_hidden_layers=24,
num_attention_heads=32,
num_key_value_heads=4,
hidden_act="silu",
max_position_embeddings=32768,
initializer_range=0.02,
rms_norm_eps=1e-6,
use_cache=True,
tie_word_embeddings=False,
rope_theta=10000.0,
rope_scaling=None,
attention_bias=False,
use_sliding_window=False,
sliding_window=4096,
max_window_layers=28,
attention_dropout=0.0,
decoder_sparse_step=1,
moe_intermediate_size=768,
num_experts_per_tok=8,
num_experts=128,
norm_topk_prob=False,
output_router_logits=False,
router_aux_loss_coef=0.001,
mlp_only_layers=None,
**kwargs,
):
self.vocab_size = vocab_size
self.max_position_embeddings = max_position_embeddings
self.hidden_size = hidden_size
self.intermediate_size = intermediate_size
self.num_hidden_layers = num_hidden_layers
self.num_attention_heads = num_attention_heads
self.use_sliding_window = use_sliding_window
self.sliding_window = sliding_window if use_sliding_window else None
self.max_window_layers = max_window_layers
self.num_key_value_heads = num_key_value_heads
self.hidden_act = hidden_act
self.initializer_range = initializer_range
self.rms_norm_eps = rms_norm_eps
self.use_cache = use_cache
self.rope_theta = rope_theta
self.rope_scaling = rope_scaling
self.attention_bias = attention_bias
self.attention_dropout = attention_dropout
# Validate the correctness of rotary position embeddings parameters
# BC: if there is a 'type' field, move it to 'rope_type'.
if self.rope_scaling is not None and "type" in self.rope_scaling:
self.rope_scaling["rope_type"] = self.rope_scaling["type"]
rope_config_validation(self)
# MoE arguments
self.decoder_sparse_step = decoder_sparse_step
self.moe_intermediate_size = moe_intermediate_size
self.num_experts_per_tok = num_experts_per_tok
self.num_experts = num_experts
self.norm_topk_prob = norm_topk_prob
self.output_router_logits = output_router_logits
self.router_aux_loss_coef = router_aux_loss_coef
self.mlp_only_layers = [] if mlp_only_layers is None else mlp_only_layers
super().__init__(
tie_word_embeddings=tie_word_embeddings,
**kwargs,
)
__all__ = ["Qwen3MoeConfig"]

1412
src/transformers/models/qwen3_moe/modeling_qwen3_moe.py Normal file
View File

File diff suppressed because it is too large Load Diff

376
src/transformers/models/qwen3_moe/modular_qwen3_moe.py Normal file
View File

@@ -0,0 +1,376 @@
# coding=utf-8
# Copyright 2025 The Qwen team, Alibaba Group and the HuggingFace Inc. team. 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.
"""PyTorch Qwen3 model."""
from typing import List, Optional, Tuple, Union
import torch
import torch.nn.functional as F
import torch.utils.checkpoint
from torch import nn
from ...activations import ACT2FN
from ...modeling_flash_attention_utils import FlashAttentionKwargs
from ...modeling_outputs import MoeCausalLMOutputWithPast
from ...processing_utils import Unpack
from ...utils import (
LossKwargs,
logging,
)
from ..llama.modeling_llama import (
LlamaForQuestionAnswering,
LlamaForSequenceClassification,
LlamaForTokenClassification,
LlamaRMSNorm,
)
from ..mixtral.modeling_mixtral import (
MixtralForCausalLM,
MixtralModel,
load_balancing_loss_func,
)
from ..qwen2_moe.modeling_qwen2_moe import Qwen2MoeDecoderLayer
from ..qwen3.modeling_qwen3 import Qwen3Attention
from .configuration_qwen3_moe import Qwen3MoeConfig
logger = logging.get_logger(__name__)
_CHECKPOINT_FOR_DOC = "Qwen/Qwen3-MoE-15B-A2B"
class Qwen3MoeAttention(Qwen3Attention): # This is the main diff with qwen2Moe!
pass
class Qwen3MoeMLP(nn.Module):
def __init__(self, config, intermediate_size=None):
super().__init__()
self.config = config
self.hidden_size = config.hidden_size
self.intermediate_size = intermediate_size if intermediate_size is not None else config.intermediate_size
self.gate_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=False)
self.up_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=False)
self.down_proj = nn.Linear(self.intermediate_size, self.hidden_size, bias=False)
self.act_fn = ACT2FN[config.hidden_act]
def forward(self, x):
down_proj = self.down_proj(self.act_fn(self.gate_proj(x)) * self.up_proj(x))
return down_proj
class Qwen3MoeSparseMoeBlock(nn.Module):
def __init__(self, config):
super().__init__()
self.num_experts = config.num_experts
self.top_k = config.num_experts_per_tok
self.norm_topk_prob = config.norm_topk_prob
# gating
self.gate = nn.Linear(config.hidden_size, config.num_experts, bias=False)
self.experts = nn.ModuleList(
[Qwen3MoeMLP(config, intermediate_size=config.moe_intermediate_size) for _ in range(self.num_experts)]
)
def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
""" """
batch_size, sequence_length, hidden_dim = hidden_states.shape
hidden_states = hidden_states.view(-1, hidden_dim)
# router_logits: (batch * sequence_length, n_experts)
router_logits = self.gate(hidden_states)
routing_weights = F.softmax(router_logits, dim=1, dtype=torch.float)
routing_weights, selected_experts = torch.topk(routing_weights, self.top_k, dim=-1)
if self.norm_topk_prob: # only diff with mixtral sparse moe block!
routing_weights /= routing_weights.sum(dim=-1, keepdim=True)
# we cast back to the input dtype
routing_weights = routing_weights.to(hidden_states.dtype)
final_hidden_states = torch.zeros(
(batch_size * sequence_length, hidden_dim), dtype=hidden_states.dtype, device=hidden_states.device
)
# One hot encode the selected experts to create an expert mask
# this will be used to easily index which expert is going to be sollicitated
expert_mask = torch.nn.functional.one_hot(selected_experts, num_classes=self.num_experts).permute(2, 1, 0)
# Loop over all available experts in the model and perform the computation on each expert
for expert_idx in range(self.num_experts):
expert_layer = self.experts[expert_idx]
idx, top_x = torch.where(expert_mask[expert_idx])
# Index the correct hidden states and compute the expert hidden state for
# the current expert. We need to make sure to multiply the output hidden
# states by `routing_weights` on the corresponding tokens (top-1 and top-2)
current_state = hidden_states[None, top_x].reshape(-1, hidden_dim)
current_hidden_states = expert_layer(current_state) * routing_weights[top_x, idx, None]
# However `index_add_` only support torch tensors for indexing so we'll use
# the `top_x` tensor here.
final_hidden_states.index_add_(0, top_x, current_hidden_states.to(hidden_states.dtype))
final_hidden_states = final_hidden_states.reshape(batch_size, sequence_length, hidden_dim)
return final_hidden_states, router_logits
class Qwen3MoeRMSNorm(LlamaRMSNorm):
pass
class Qwen3MoeDecoderLayer(Qwen2MoeDecoderLayer, nn.Module):
def __init__(self, config: Qwen3MoeConfig, layer_idx: int):
nn.Module().__init__()
self.hidden_size = config.hidden_size
self.self_attn = Qwen3MoeAttention(config, layer_idx)
self.mlp = Qwen3MoeMLP(config)
self.self_attn = Qwen3MoeAttention(config, layer_idx)
if (layer_idx not in config.mlp_only_layers) and (
config.num_experts > 0 and (layer_idx + 1) % config.decoder_sparse_step == 0
):
self.mlp = Qwen3MoeSparseMoeBlock(config)
else:
self.mlp = Qwen3MoeMLP(config, intermediate_size=config.intermediate_size)
self.input_layernorm = Qwen3MoeRMSNorm(config.hidden_size, eps=config.rms_norm_eps)
self.post_attention_layernorm = Qwen3MoeRMSNorm(config.hidden_size, eps=config.rms_norm_eps)
def forward(
self,
hidden_states: torch.Tensor,
attention_mask: Optional[torch.Tensor] = None,
position_ids: Optional[torch.LongTensor] = None,
past_key_value: Optional[Tuple[torch.Tensor]] = None,
output_attentions: Optional[bool] = False,
output_router_logits: Optional[bool] = False,
use_cache: Optional[bool] = False,
cache_position: Optional[torch.LongTensor] = None,
position_embeddings: Optional[Tuple[torch.Tensor, torch.Tensor]] = None, # necessary, but kept here for BC
**kwargs: Unpack[FlashAttentionKwargs],
) -> Tuple[torch.FloatTensor, Optional[Tuple[torch.FloatTensor, torch.FloatTensor]]]:
"""
Args:
hidden_states (`torch.FloatTensor`): input to the layer of shape `(batch, seq_len, embed_dim)`
attention_mask (`torch.FloatTensor`, *optional*): attention mask of size
`(batch, sequence_length)` where padding elements are indicated by 0.
output_attentions (`bool`, *optional*):
Whether or not to return the attentions tensors of all attention layers. See `attentions` under
returned tensors for more detail.
output_router_logits (`bool`, *optional*):
Whether or not to return the logits of all the routers. They are useful for computing the router loss,
and should not be returned during inference.
use_cache (`bool`, *optional*):
If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding
(see `past_key_values`).
past_key_value (`Tuple(torch.FloatTensor)`, *optional*): cached past key and value projection states
cache_position (`torch.LongTensor` of shape `(sequence_length)`, *optional*):
Indices depicting the position of the input sequence tokens in the sequence.
position_embeddings (`Tuple[torch.FloatTensor, torch.FloatTensor]`, *optional*):
Tuple containing the cosine and sine positional embeddings of shape `(batch_size, seq_len, head_dim)`,
with `head_dim` being the embedding dimension of each attention head.
kwargs (`dict`, *optional*):
Arbitrary kwargs to be ignored, used for FSDP and other methods that injects code
into the model
"""
residual = hidden_states
hidden_states = self.input_layernorm(hidden_states)
# Self Attention
hidden_states, self_attn_weights = self.self_attn(
hidden_states=hidden_states,
attention_mask=attention_mask,
position_ids=position_ids,
past_key_value=past_key_value,
output_attentions=output_attentions,
use_cache=use_cache,
cache_position=cache_position,
position_embeddings=position_embeddings,
)
hidden_states = residual + hidden_states
# Fully Connected
residual = hidden_states
hidden_states = self.post_attention_layernorm(hidden_states)
hidden_states = self.mlp(hidden_states)
if isinstance(hidden_states, tuple):
hidden_states, router_logits = hidden_states
else:
router_logits = None
hidden_states = residual + hidden_states
outputs = (hidden_states,)
if output_attentions:
outputs += (self_attn_weights,)
if output_router_logits:
outputs += (router_logits,)
return outputs
class Qwen3MoeModel(MixtralModel):
def __init__(self, config: Qwen3MoeConfig):
super().__init__(config)
self.layers = nn.ModuleList(
[Qwen3MoeDecoderLayer(config, layer_idx) for layer_idx in range(config.num_hidden_layers)]
)
class KwargsForCausalLM(FlashAttentionKwargs, LossKwargs): ...
class Qwen3MoeForCausalLM(MixtralForCausalLM):
def __init__(self, config):
super().__init__(config)
self.model = Qwen3MoeModel(config)
self.num_experts = config.num_experts
def forward(
self,
input_ids: torch.LongTensor = None,
attention_mask: Optional[torch.Tensor] = None,
position_ids: Optional[torch.LongTensor] = None,
past_key_values: Optional[List[torch.FloatTensor]] = None,
inputs_embeds: Optional[torch.FloatTensor] = None,
labels: Optional[torch.LongTensor] = None,
use_cache: Optional[bool] = None,
output_attentions: Optional[bool] = None,
output_hidden_states: Optional[bool] = None,
output_router_logits: Optional[bool] = None,
return_dict: Optional[bool] = None,
cache_position: Optional[torch.LongTensor] = None,
logits_to_keep: Union[int, torch.Tensor] = 0,
**kwargs: Unpack[KwargsForCausalLM],
) -> Union[Tuple, MoeCausalLMOutputWithPast]:
r"""
labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
Labels for computing the masked language modeling loss. Indices should either be in `[0, ...,
config.vocab_size]` or -100 (see `input_ids` docstring). Tokens with indices set to `-100` are ignored
(masked), the loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`.
logits_to_keep (`int` or `torch.Tensor`, *optional*):
If an `int`, compute logits for the last `logits_to_keep` tokens. If `0`, calculate logits for all
`input_ids` (special case). Only last token logits are needed for generation, and calculating them only for that
token can save memory, which becomes pretty significant for long sequences or large vocabulary size.
If a `torch.Tensor`, must be 1D corresponding to the indices to keep in the sequence length dimension.
This is useful when using packed tensor format (single dimension for batch and sequence length).
Returns:
Example:
```python
>>> from transformers import AutoTokenizer, Qwen3MoeForCausalLM
>>> model = Qwen3MoeForCausalLM.from_pretrained("Qwen/Qwen3-MoE-15B-A2B")
>>> tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen3-MoE-15B-A2B")
>>> prompt = "Hey, are you conscious? Can you talk to me?"
>>> inputs = tokenizer(prompt, return_tensors="pt")
>>> # Generate
>>> generate_ids = model.generate(inputs.input_ids, max_length=30)
>>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
"Hey, are you conscious? Can you talk to me?\nI'm not conscious, but I can talk to you."
```"""
output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
output_router_logits = (
output_router_logits if output_router_logits is not None else self.config.output_router_logits
)
output_hidden_states = (
output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
)
return_dict = return_dict if return_dict is not None else self.config.use_return_dict
# decoder outputs consists of (dec_features, layer_state, dec_hidden, dec_attn)
outputs = self.model(
input_ids=input_ids,
attention_mask=attention_mask,
position_ids=position_ids,
past_key_values=past_key_values,
inputs_embeds=inputs_embeds,
use_cache=use_cache,
output_attentions=output_attentions,
output_hidden_states=output_hidden_states,
output_router_logits=output_router_logits,
return_dict=return_dict,
cache_position=cache_position,
**kwargs,
)
hidden_states = outputs[0]
# Only compute necessary logits, and do not upcast them to float if we are not computing the loss
slice_indices = slice(-logits_to_keep, None) if isinstance(logits_to_keep, int) else logits_to_keep
logits = self.lm_head(hidden_states[:, slice_indices, :])
loss = None
if labels is not None:
loss = self.loss_function(logits, labels, self.vocab_size, **kwargs)
aux_loss = None
if output_router_logits:
aux_loss = load_balancing_loss_func(
outputs.router_logits if return_dict else outputs[-1],
self.num_experts,
self.num_experts_per_tok,
attention_mask,
)
if labels is not None:
loss += self.router_aux_loss_coef * aux_loss.to(loss.device) # make sure to reside in the same device
if not return_dict:
output = (logits,) + outputs[1:]
if output_router_logits:
output = (aux_loss,) + output
return (loss,) + output if loss is not None else output
return MoeCausalLMOutputWithPast(
loss=loss,
aux_loss=aux_loss,
logits=logits,
past_key_values=outputs.past_key_values,
hidden_states=outputs.hidden_states,
attentions=outputs.attentions,
router_logits=outputs.router_logits,
)
class Qwen3MoeForSequenceClassification(LlamaForSequenceClassification):
pass
class Qwen3MoeForTokenClassification(LlamaForTokenClassification):
pass
class Qwen3MoeForQuestionAnswering(LlamaForQuestionAnswering):
pass
__all__ = [
"Qwen3MoeForCausalLM",
"Qwen3MoeForQuestionAnswering",
"Qwen3MoeModel",
"Qwen3MoePreTrainedModel", # noqa: F822
"Qwen3MoeForSequenceClassification",
"Qwen3MoeForTokenClassification",
]

84
src/transformers/utils/dummy_pt_objects.py
View File

@@ -8222,6 +8222,90 @@ class Qwen2VLPreTrainedModel(metaclass=DummyObject):
requires_backends(self, ["torch"])
class Qwen3ForCausalLM(metaclass=DummyObject):
_backends = ["torch"]
def __init__(self, *args, **kwargs):
requires_backends(self, ["torch"])
class Qwen3ForQuestionAnswering(metaclass=DummyObject):
_backends = ["torch"]
def __init__(self, *args, **kwargs):
requires_backends(self, ["torch"])
class Qwen3ForSequenceClassification(metaclass=DummyObject):
_backends = ["torch"]
def __init__(self, *args, **kwargs):
requires_backends(self, ["torch"])
class Qwen3ForTokenClassification(metaclass=DummyObject):
_backends = ["torch"]
def __init__(self, *args, **kwargs):
requires_backends(self, ["torch"])
class Qwen3Model(metaclass=DummyObject):
_backends = ["torch"]
def __init__(self, *args, **kwargs):
requires_backends(self, ["torch"])
class Qwen3PreTrainedModel(metaclass=DummyObject):
_backends = ["torch"]
def __init__(self, *args, **kwargs):
requires_backends(self, ["torch"])
class Qwen3MoeForCausalLM(metaclass=DummyObject):
_backends = ["torch"]
def __init__(self, *args, **kwargs):
requires_backends(self, ["torch"])
class Qwen3MoeForQuestionAnswering(metaclass=DummyObject):
_backends = ["torch"]
def __init__(self, *args, **kwargs):
requires_backends(self, ["torch"])
class Qwen3MoeForSequenceClassification(metaclass=DummyObject):
_backends = ["torch"]
def __init__(self, *args, **kwargs):
requires_backends(self, ["torch"])
class Qwen3MoeForTokenClassification(metaclass=DummyObject):
_backends = ["torch"]
def __init__(self, *args, **kwargs):
requires_backends(self, ["torch"])
class Qwen3MoeModel(metaclass=DummyObject):
_backends = ["torch"]
def __init__(self, *args, **kwargs):
requires_backends(self, ["torch"])
class Qwen3MoePreTrainedModel(metaclass=DummyObject):
_backends = ["torch"]
def __init__(self, *args, **kwargs):
requires_backends(self, ["torch"])
class RagModel(metaclass=DummyObject):
_backends = ["torch"]

2
src/transformers/utils/fx.py
View File

@@ -158,6 +158,8 @@ _REGULAR_SUPPORTED_MODEL_NAMES_AND_TASKS = [
"plbart",
"qwen2",
"qwen2_moe",
"qwen3",
"qwen3_moe",
"resnet",
"roberta",
"segformer",