transformers 4.57.2__py3-none-any.whl → 4.57.4__py3-none-any.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (119) hide show
  1. transformers/__init__.py +1 -1
  2. transformers/generation/utils.py +4 -2
  3. transformers/models/apertus/modeling_apertus.py +1 -1
  4. transformers/models/arcee/modeling_arcee.py +1 -1
  5. transformers/models/aria/modeling_aria.py +1 -1
  6. transformers/models/audio_spectrogram_transformer/modeling_audio_spectrogram_transformer.py +1 -1
  7. transformers/models/aya_vision/modeling_aya_vision.py +1 -1
  8. transformers/models/aya_vision/modular_aya_vision.py +1 -1
  9. transformers/models/bitnet/modeling_bitnet.py +1 -1
  10. transformers/models/blip_2/modeling_blip_2.py +1 -1
  11. transformers/models/blt/modeling_blt.py +2 -2
  12. transformers/models/blt/modular_blt.py +2 -2
  13. transformers/models/cohere/modeling_cohere.py +1 -1
  14. transformers/models/cohere2/modeling_cohere2.py +1 -1
  15. transformers/models/cohere2_vision/modeling_cohere2_vision.py +2 -2
  16. transformers/models/cohere2_vision/modular_cohere2_vision.py +2 -2
  17. transformers/models/csm/modeling_csm.py +2 -2
  18. transformers/models/csm/modular_csm.py +2 -2
  19. transformers/models/deepseek_v2/modeling_deepseek_v2.py +1 -1
  20. transformers/models/deepseek_v3/modeling_deepseek_v3.py +1 -1
  21. transformers/models/diffllama/modeling_diffllama.py +1 -1
  22. transformers/models/dinov2/modeling_dinov2.py +1 -1
  23. transformers/models/dinov2_with_registers/modeling_dinov2_with_registers.py +1 -1
  24. transformers/models/doge/modeling_doge.py +1 -1
  25. transformers/models/dots1/modeling_dots1.py +1 -1
  26. transformers/models/edgetam/modeling_edgetam.py +2 -2
  27. transformers/models/edgetam/modular_edgetam.py +1 -1
  28. transformers/models/efficientloftr/modeling_efficientloftr.py +1 -1
  29. transformers/models/emu3/modeling_emu3.py +1 -1
  30. transformers/models/eomt/modeling_eomt.py +1 -1
  31. transformers/models/eomt/modular_eomt.py +1 -1
  32. transformers/models/ernie4_5/modeling_ernie4_5.py +1 -1
  33. transformers/models/ernie4_5_moe/modeling_ernie4_5_moe.py +1 -1
  34. transformers/models/ernie4_5_moe/modular_ernie4_5_moe.py +1 -1
  35. transformers/models/esm/modeling_esm.py +1 -1
  36. transformers/models/evolla/modeling_evolla.py +2 -2
  37. transformers/models/evolla/modular_evolla.py +2 -2
  38. transformers/models/exaone4/modeling_exaone4.py +1 -1
  39. transformers/models/exaone4/modular_exaone4.py +1 -1
  40. transformers/models/flex_olmo/modeling_flex_olmo.py +1 -1
  41. transformers/models/flex_olmo/modular_flex_olmo.py +1 -1
  42. transformers/models/gemma/modeling_gemma.py +1 -1
  43. transformers/models/gemma2/modeling_gemma2.py +1 -1
  44. transformers/models/gemma3/modeling_gemma3.py +1 -1
  45. transformers/models/glm/modeling_glm.py +1 -1
  46. transformers/models/glm4/modeling_glm4.py +1 -1
  47. transformers/models/glm4_moe/modeling_glm4_moe.py +1 -1
  48. transformers/models/glm4v/modeling_glm4v.py +1 -1
  49. transformers/models/glm4v/modular_glm4v.py +1 -1
  50. transformers/models/glm4v_moe/modeling_glm4v_moe.py +1 -1
  51. transformers/models/gpt_neox/modeling_gpt_neox.py +1 -1
  52. transformers/models/gpt_oss/modeling_gpt_oss.py +1 -1
  53. transformers/models/gpt_oss/modular_gpt_oss.py +1 -1
  54. transformers/models/granite/modeling_granite.py +1 -1
  55. transformers/models/helium/modeling_helium.py +1 -1
  56. transformers/models/hunyuan_v1_dense/modeling_hunyuan_v1_dense.py +1 -1
  57. transformers/models/hunyuan_v1_moe/modeling_hunyuan_v1_moe.py +1 -1
  58. transformers/models/idefics/modeling_idefics.py +1 -1
  59. transformers/models/instructblip/modeling_instructblip.py +1 -1
  60. transformers/models/instructblipvideo/modeling_instructblipvideo.py +1 -1
  61. transformers/models/lfm2/modeling_lfm2.py +1 -1
  62. transformers/models/llama/modeling_llama.py +1 -1
  63. transformers/models/llama4/modeling_llama4.py +1 -1
  64. transformers/models/longcat_flash/modeling_longcat_flash.py +1 -1
  65. transformers/models/minimax/modeling_minimax.py +1 -1
  66. transformers/models/minimax/modular_minimax.py +1 -1
  67. transformers/models/ministral/modeling_ministral.py +1 -1
  68. transformers/models/ministral/modular_ministral.py +1 -1
  69. transformers/models/mistral/modeling_mistral.py +1 -1
  70. transformers/models/mistral/modular_mistral.py +1 -1
  71. transformers/models/mixtral/modeling_mixtral.py +1 -1
  72. transformers/models/mllama/modeling_mllama.py +3 -3
  73. transformers/models/modernbert_decoder/modeling_modernbert_decoder.py +1 -1
  74. transformers/models/modernbert_decoder/modular_modernbert_decoder.py +1 -1
  75. transformers/models/moonshine/modeling_moonshine.py +2 -2
  76. transformers/models/moonshine/modular_moonshine.py +2 -2
  77. transformers/models/olmo/modeling_olmo.py +1 -1
  78. transformers/models/olmo2/modeling_olmo2.py +1 -1
  79. transformers/models/olmo3/modeling_olmo3.py +1 -1
  80. transformers/models/parakeet/modeling_parakeet.py +1 -1
  81. transformers/models/parakeet/modular_parakeet.py +1 -1
  82. transformers/models/phi/modeling_phi.py +1 -1
  83. transformers/models/phi3/modeling_phi3.py +1 -1
  84. transformers/models/phi4_multimodal/modeling_phi4_multimodal.py +1 -1
  85. transformers/models/phi4_multimodal/modular_phi4_multimodal.py +1 -1
  86. transformers/models/qwen2/modeling_qwen2.py +1 -1
  87. transformers/models/qwen2/modular_qwen2.py +1 -1
  88. transformers/models/qwen3/modeling_qwen3.py +1 -1
  89. transformers/models/qwen3_moe/modeling_qwen3_moe.py +1 -1
  90. transformers/models/qwen3_next/modeling_qwen3_next.py +1 -1
  91. transformers/models/qwen3_next/modular_qwen3_next.py +1 -1
  92. transformers/models/qwen3_omni_moe/modeling_qwen3_omni_moe.py +4 -4
  93. transformers/models/qwen3_omni_moe/modular_qwen3_omni_moe.py +1 -1
  94. transformers/models/qwen3_vl/modeling_qwen3_vl.py +3 -3
  95. transformers/models/qwen3_vl/modular_qwen3_vl.py +3 -3
  96. transformers/models/qwen3_vl_moe/modeling_qwen3_vl_moe.py +3 -3
  97. transformers/models/sam/modeling_sam.py +1 -1
  98. transformers/models/sam2/modeling_sam2.py +3 -3
  99. transformers/models/sam2/modular_sam2.py +3 -3
  100. transformers/models/sam_hq/modeling_sam_hq.py +1 -1
  101. transformers/models/seed_oss/modeling_seed_oss.py +1 -1
  102. transformers/models/siglip/modeling_siglip.py +1 -1
  103. transformers/models/siglip2/modeling_siglip2.py +1 -1
  104. transformers/models/smollm3/modeling_smollm3.py +1 -1
  105. transformers/models/starcoder2/modeling_starcoder2.py +1 -1
  106. transformers/models/starcoder2/modular_starcoder2.py +1 -1
  107. transformers/models/t5gemma/modeling_t5gemma.py +2 -2
  108. transformers/models/t5gemma/modular_t5gemma.py +2 -2
  109. transformers/models/vaultgemma/modeling_vaultgemma.py +1 -1
  110. transformers/models/voxtral/modeling_voxtral.py +1 -1
  111. transformers/models/voxtral/modular_voxtral.py +1 -1
  112. transformers/tokenization_utils_base.py +88 -32
  113. transformers/utils/generic.py +3 -1
  114. {transformers-4.57.2.dist-info → transformers-4.57.4.dist-info}/METADATA +1 -1
  115. {transformers-4.57.2.dist-info → transformers-4.57.4.dist-info}/RECORD +119 -119
  116. {transformers-4.57.2.dist-info → transformers-4.57.4.dist-info}/WHEEL +0 -0
  117. {transformers-4.57.2.dist-info → transformers-4.57.4.dist-info}/entry_points.txt +0 -0
  118. {transformers-4.57.2.dist-info → transformers-4.57.4.dist-info}/licenses/LICENSE +0 -0
  119. {transformers-4.57.2.dist-info → transformers-4.57.4.dist-info}/top_level.txt +0 -0
transformers/models/hunyuan_v1_dense/modeling_hunyuan_v1_dense.py CHANGED
@@ -365,7 +365,7 @@ class HunYuanDenseV1Model(HunYuanDenseV1PreTrainedModel):
365
365
  # Initialize weights and apply final processing
366
366
  self.post_init()
367
367
 
368
- @check_model_inputs()
368
+ @check_model_inputs
369
369
  @auto_docstring
370
370
  def forward(
371
371
  self,
transformers/models/hunyuan_v1_moe/modeling_hunyuan_v1_moe.py CHANGED
@@ -435,7 +435,7 @@ class HunYuanMoEV1Model(HunYuanMoEV1PreTrainedModel):
435
435
  # Initialize weights and apply final processing
436
436
  self.post_init()
437
437
 
438
- @check_model_inputs()
438
+ @check_model_inputs
439
439
  @auto_docstring
440
440
  def forward(
441
441
  self,
transformers/models/idefics/modeling_idefics.py CHANGED
@@ -943,7 +943,7 @@ class IdeficsModel(IdeficsPreTrainedModel):
943
943
  def freeze_vision_layers(self, module_exceptions=[]):
944
944
  freeze_model(self.vision_model, module_exceptions=module_exceptions)
945
945
 
946
- @check_model_inputs()
946
+ @check_model_inputs
947
947
  @auto_docstring
948
948
  def forward(
949
949
  self,
transformers/models/instructblip/modeling_instructblip.py CHANGED
@@ -915,7 +915,7 @@ class InstructBlipQFormerModel(InstructBlipPreTrainedModel):
915
915
  extended_attention_mask = (1.0 - extended_attention_mask) * -10000.0
916
916
  return extended_attention_mask
917
917
 
918
- @check_model_inputs()
918
+ @check_model_inputs
919
919
  @auto_docstring
920
920
  def forward(
921
921
  self,
transformers/models/instructblipvideo/modeling_instructblipvideo.py CHANGED
@@ -877,7 +877,7 @@ class InstructBlipVideoQFormerModel(InstructBlipVideoPreTrainedModel):
877
877
  extended_attention_mask = (1.0 - extended_attention_mask) * -10000.0
878
878
  return extended_attention_mask
879
879
 
880
- @check_model_inputs()
880
+ @check_model_inputs
881
881
  @auto_docstring
882
882
  def forward(
883
883
  self,
transformers/models/lfm2/modeling_lfm2.py CHANGED
@@ -609,7 +609,7 @@ class Lfm2Model(Lfm2PreTrainedModel):
609
609
  # Initialize weights and apply final processing
610
610
  self.post_init()
611
611
 
612
- @check_model_inputs()
612
+ @check_model_inputs
613
613
  @auto_docstring
614
614
  def forward(
615
615
  self,
transformers/models/llama/modeling_llama.py CHANGED
@@ -348,7 +348,7 @@ class LlamaModel(LlamaPreTrainedModel):
348
348
  # Initialize weights and apply final processing
349
349
  self.post_init()
350
350
 
351
- @check_model_inputs()
351
+ @check_model_inputs
352
352
  @auto_docstring
353
353
  def forward(
354
354
  self,
transformers/models/llama4/modeling_llama4.py CHANGED
@@ -491,7 +491,7 @@ class Llama4TextModel(Llama4PreTrainedModel):
491
491
  self.post_init()
492
492
 
493
493
  @can_return_tuple
494
- @check_model_inputs()
494
+ @check_model_inputs
495
495
  @auto_docstring
496
496
  def forward(
497
497
  self,
transformers/models/longcat_flash/modeling_longcat_flash.py CHANGED
@@ -541,7 +541,7 @@ class LongcatFlashModel(LongcatFlashPreTrainedModel):
541
541
  # Initialize weights and apply final processing
542
542
  self.post_init()
543
543
 
544
- @check_model_inputs()
544
+ @check_model_inputs
545
545
  @auto_docstring
546
546
  def forward(
547
547
  self,
transformers/models/minimax/modeling_minimax.py CHANGED
@@ -646,7 +646,7 @@ class MiniMaxModel(MiniMaxPreTrainedModel):
646
646
  # Initialize weights and apply final processing
647
647
  self.post_init()
648
648
 
649
- @check_model_inputs()
649
+ @check_model_inputs
650
650
  def forward(
651
651
  self,
652
652
  input_ids: Optional[torch.LongTensor] = None,
transformers/models/minimax/modular_minimax.py CHANGED
@@ -480,7 +480,7 @@ class MiniMaxPreTrainedModel(MixtralPreTrainedModel):
480
480
 
481
481
 
482
482
  class MiniMaxModel(MixtralModel):
483
- @check_model_inputs()
483
+ @check_model_inputs
484
484
  def forward(
485
485
  self,
486
486
  input_ids: Optional[torch.LongTensor] = None,
transformers/models/ministral/modeling_ministral.py CHANGED
@@ -325,7 +325,7 @@ class MinistralModel(MinistralPreTrainedModel):
325
325
  # Initialize weights and apply final processing
326
326
  self.post_init()
327
327
 
328
- @check_model_inputs()
328
+ @check_model_inputs
329
329
  @auto_docstring
330
330
  def forward(
331
331
  self,
transformers/models/ministral/modular_ministral.py CHANGED
@@ -198,7 +198,7 @@ class MinistralModel(Qwen2Model):
198
198
  super().__init__(config)
199
199
  del self.has_sliding_layers
200
200
 
201
- @check_model_inputs()
201
+ @check_model_inputs
202
202
  @auto_docstring
203
203
  def forward(
204
204
  self,
transformers/models/mistral/modeling_mistral.py CHANGED
@@ -321,7 +321,7 @@ class MistralModel(MistralPreTrainedModel):
321
321
  # Initialize weights and apply final processing
322
322
  self.post_init()
323
323
 
324
- @check_model_inputs()
324
+ @check_model_inputs
325
325
  @auto_docstring
326
326
  def forward(
327
327
  self,
transformers/models/mistral/modular_mistral.py CHANGED
@@ -112,7 +112,7 @@ class MistralPreTrainedModel(LlamaPreTrainedModel):
112
112
 
113
113
 
114
114
  class MistralModel(LlamaModel):
115
- @check_model_inputs()
115
+ @check_model_inputs
116
116
  @auto_docstring
117
117
  def forward(
118
118
  self,
transformers/models/mixtral/modeling_mixtral.py CHANGED
@@ -415,7 +415,7 @@ class MixtralModel(MixtralPreTrainedModel):
415
415
  # Initialize weights and apply final processing
416
416
  self.post_init()
417
417
 
418
- @check_model_inputs()
418
+ @check_model_inputs
419
419
  @auto_docstring
420
420
  def forward(
421
421
  self,
transformers/models/mllama/modeling_mllama.py CHANGED
@@ -993,7 +993,7 @@ class MllamaVisionModel(MllamaPreTrainedModel):
993
993
  hidden_state = torch.cat([class_embedding, hidden_state], dim=1)
994
994
  return hidden_state
995
995
 
996
- @check_model_inputs()
996
+ @check_model_inputs
997
997
  @auto_docstring
998
998
  def forward(
999
999
  self, pixel_values: torch.Tensor, aspect_ratio_ids: torch.Tensor, aspect_ratio_mask: torch.Tensor, **kwargs
@@ -1161,7 +1161,7 @@ class MllamaTextModel(MllamaPreTrainedModel):
1161
1161
  self.gradient_checkpointing = False
1162
1162
  self.post_init()
1163
1163
 
1164
- @check_model_inputs()
1164
+ @check_model_inputs
1165
1165
  @can_return_tuple
1166
1166
  @auto_docstring
1167
1167
  def forward(
@@ -1429,7 +1429,7 @@ class MllamaModel(MllamaPreTrainedModel):
1429
1429
  def get_decoder(self):
1430
1430
  return self.language_model
1431
1431
 
1432
- @check_model_inputs()
1432
+ @check_model_inputs
1433
1433
  @can_return_tuple
1434
1434
  @auto_docstring
1435
1435
  def forward(
transformers/models/modernbert_decoder/modeling_modernbert_decoder.py CHANGED
@@ -430,7 +430,7 @@ class ModernBertDecoderModel(ModernBertDecoderPreTrainedModel):
430
430
  def set_input_embeddings(self, value):
431
431
  self.embeddings.tok_embeddings = value
432
432
 
433
- @check_model_inputs()
433
+ @check_model_inputs
434
434
  @auto_docstring
435
435
  def forward(
436
436
  self,
transformers/models/modernbert_decoder/modular_modernbert_decoder.py CHANGED
@@ -501,7 +501,7 @@ class ModernBertDecoderModel(ModernBertDecoderPreTrainedModel):
501
501
  def set_input_embeddings(self, value):
502
502
  self.embeddings.tok_embeddings = value
503
503
 
504
- @check_model_inputs()
504
+ @check_model_inputs
505
505
  @auto_docstring
506
506
  def forward(
507
507
  self,
transformers/models/moonshine/modeling_moonshine.py CHANGED
@@ -520,7 +520,7 @@ class MoonshineEncoder(MoonshinePreTrainedModel):
520
520
  def set_input_embeddings(self, value: nn.Module):
521
521
  self.conv1 = value
522
522
 
523
- @check_model_inputs()
523
+ @check_model_inputs
524
524
  def forward(
525
525
  self,
526
526
  input_values: torch.FloatTensor,
@@ -605,7 +605,7 @@ class MoonshineDecoder(MoonshinePreTrainedModel):
605
605
  # Initialize weights and apply final processing
606
606
  self.post_init()
607
607
 
608
- @check_model_inputs()
608
+ @check_model_inputs
609
609
  def forward(
610
610
  self,
611
611
  input_ids: Optional[torch.LongTensor] = None,
transformers/models/moonshine/modular_moonshine.py CHANGED
@@ -552,7 +552,7 @@ class MoonshineEncoder(MoonshinePreTrainedModel):
552
552
  def set_input_embeddings(self, value: nn.Module):
553
553
  self.conv1 = value
554
554
 
555
- @check_model_inputs()
555
+ @check_model_inputs
556
556
  def forward(
557
557
  self,
558
558
  input_values: torch.FloatTensor,
@@ -627,7 +627,7 @@ class MoonshineDecoder(LlamaModel):
627
627
  [MoonshineDecoderLayer(config, idx) for idx in range(config.decoder_num_hidden_layers)]
628
628
  )
629
629
 
630
- @check_model_inputs()
630
+ @check_model_inputs
631
631
  def forward(
632
632
  self,
633
633
  input_ids: Optional[torch.LongTensor] = None,
transformers/models/olmo/modeling_olmo.py CHANGED
@@ -325,7 +325,7 @@ class OlmoModel(OlmoPreTrainedModel):
325
325
  # Initialize weights and apply final processing
326
326
  self.post_init()
327
327
 
328
- @check_model_inputs()
328
+ @check_model_inputs
329
329
  @auto_docstring
330
330
  def forward(
331
331
  self,
transformers/models/olmo2/modeling_olmo2.py CHANGED
@@ -330,7 +330,7 @@ class Olmo2Model(Olmo2PreTrainedModel):
330
330
  # Initialize weights and apply final processing
331
331
  self.post_init()
332
332
 
333
- @check_model_inputs()
333
+ @check_model_inputs
334
334
  @auto_docstring
335
335
  def forward(
336
336
  self,
transformers/models/olmo3/modeling_olmo3.py CHANGED
@@ -358,7 +358,7 @@ class Olmo3Model(Olmo3PreTrainedModel):
358
358
  # Initialize weights and apply final processing
359
359
  self.post_init()
360
360
 
361
- @check_model_inputs()
361
+ @check_model_inputs
362
362
  @auto_docstring
363
363
  def forward(
364
364
  self,
transformers/models/parakeet/modeling_parakeet.py CHANGED
@@ -507,7 +507,7 @@ class ParakeetEncoder(ParakeetPreTrainedModel):
507
507
  self.post_init()
508
508
 
509
509
  @auto_docstring
510
- @check_model_inputs()
510
+ @check_model_inputs
511
511
  @can_return_tuple
512
512
  def forward(
513
513
  self,
transformers/models/parakeet/modular_parakeet.py CHANGED
@@ -391,7 +391,7 @@ class ParakeetEncoder(ParakeetPreTrainedModel):
391
391
  self.post_init()
392
392
 
393
393
  @auto_docstring
394
- @check_model_inputs()
394
+ @check_model_inputs
395
395
  @can_return_tuple
396
396
  def forward(
397
397
  self,
transformers/models/phi/modeling_phi.py CHANGED
@@ -330,7 +330,7 @@ class PhiModel(PhiPreTrainedModel):
330
330
  # Initialize weights and apply final processing
331
331
  self.post_init()
332
332
 
333
- @check_model_inputs()
333
+ @check_model_inputs
334
334
  @auto_docstring
335
335
  def forward(
336
336
  self,
transformers/models/phi3/modeling_phi3.py CHANGED
@@ -353,7 +353,7 @@ class Phi3Model(Phi3PreTrainedModel):
353
353
  # Initialize weights and apply final processing
354
354
  self.post_init()
355
355
 
356
- @check_model_inputs()
356
+ @check_model_inputs
357
357
  @auto_docstring
358
358
  def forward(
359
359
  self,
transformers/models/phi4_multimodal/modeling_phi4_multimodal.py CHANGED
@@ -1562,7 +1562,7 @@ class Phi4MultimodalModel(Phi4MultimodalPreTrainedModel):
1562
1562
  # Initialize weights and apply final processing
1563
1563
  self.post_init()
1564
1564
 
1565
- @check_model_inputs()
1565
+ @check_model_inputs
1566
1566
  def forward(
1567
1567
  self,
1568
1568
  input_ids: Optional[torch.LongTensor] = None,
transformers/models/phi4_multimodal/modular_phi4_multimodal.py CHANGED
@@ -1472,7 +1472,7 @@ class Phi4MultimodalModel(Phi3Model):
1472
1472
  # Initialize weights and apply final processing
1473
1473
  self.post_init()
1474
1474
 
1475
- @check_model_inputs()
1475
+ @check_model_inputs
1476
1476
  def forward(
1477
1477
  self,
1478
1478
  input_ids: Optional[torch.LongTensor] = None,
transformers/models/qwen2/modeling_qwen2.py CHANGED
@@ -325,7 +325,7 @@ class Qwen2Model(Qwen2PreTrainedModel):
325
325
  # Initialize weights and apply final processing
326
326
  self.post_init()
327
327
 
328
- @check_model_inputs()
328
+ @check_model_inputs
329
329
  @auto_docstring
330
330
  def forward(
331
331
  self,
transformers/models/qwen2/modular_qwen2.py CHANGED
@@ -143,7 +143,7 @@ class Qwen2Model(MistralModel):
143
143
  super().__init__(config)
144
144
  self.has_sliding_layers = "sliding_attention" in self.config.layer_types
145
145
 
146
- @check_model_inputs()
146
+ @check_model_inputs
147
147
  @auto_docstring
148
148
  def forward(
149
149
  self,
transformers/models/qwen3/modeling_qwen3.py CHANGED
@@ -351,7 +351,7 @@ class Qwen3Model(Qwen3PreTrainedModel):
351
351
  # Initialize weights and apply final processing
352
352
  self.post_init()
353
353
 
354
- @check_model_inputs()
354
+ @check_model_inputs
355
355
  @auto_docstring
356
356
  def forward(
357
357
  self,
transformers/models/qwen3_moe/modeling_qwen3_moe.py CHANGED
@@ -438,7 +438,7 @@ class Qwen3MoeModel(Qwen3MoePreTrainedModel):
438
438
  # Initialize weights and apply final processing
439
439
  self.post_init()
440
440
 
441
- @check_model_inputs()
441
+ @check_model_inputs
442
442
  @auto_docstring
443
443
  def forward(
444
444
  self,
transformers/models/qwen3_next/modeling_qwen3_next.py CHANGED
@@ -988,7 +988,7 @@ class Qwen3NextModel(Qwen3NextPreTrainedModel):
988
988
  # Initialize weights and apply final processing
989
989
  self.post_init()
990
990
 
991
- @check_model_inputs()
991
+ @check_model_inputs
992
992
  @auto_docstring
993
993
  def forward(
994
994
  self,
transformers/models/qwen3_next/modular_qwen3_next.py CHANGED
@@ -727,7 +727,7 @@ class Qwen3NextModel(Qwen3NextPreTrainedModel):
727
727
  # Initialize weights and apply final processing
728
728
  self.post_init()
729
729
 
730
- @check_model_inputs()
730
+ @check_model_inputs
731
731
  @auto_docstring
732
732
  def forward(
733
733
  self,
transformers/models/qwen3_omni_moe/modeling_qwen3_omni_moe.py CHANGED
@@ -1628,7 +1628,7 @@ class Qwen3OmniMoeThinkerTextModel(Qwen3OmniMoePreTrainedModel):
1628
1628
  # Initialize weights and apply final processing
1629
1629
  self.post_init()
1630
1630
 
1631
- @check_model_inputs()
1631
+ @check_model_inputs
1632
1632
  @auto_docstring
1633
1633
  def forward(
1634
1634
  self,
@@ -2480,7 +2480,7 @@ class Qwen3OmniMoeTalkerCodePredictorModel(Qwen3OmniMoePreTrainedModel):
2480
2480
  # Initialize weights and apply final processing
2481
2481
  self.post_init()
2482
2482
 
2483
- @check_model_inputs()
2483
+ @check_model_inputs
2484
2484
  @auto_docstring
2485
2485
  def forward(
2486
2486
  self,
@@ -2852,7 +2852,7 @@ class Qwen3OmniMoeTalkerModel(Qwen3OmniMoePreTrainedModel):
2852
2852
  # Initialize weights and apply final processing
2853
2853
  self.post_init()
2854
2854
 
2855
- @check_model_inputs()
2855
+ @check_model_inputs
2856
2856
  @auto_docstring
2857
2857
  def forward(
2858
2858
  self,
@@ -3542,7 +3542,7 @@ class Qwen3OmniMoeCode2WavTransformerModel(Qwen3OmniMoePreTrainedModel):
3542
3542
  # Initialize weights and apply final processing
3543
3543
  self.post_init()
3544
3544
 
3545
- @check_model_inputs()
3545
+ @check_model_inputs
3546
3546
  @auto_docstring
3547
3547
  def forward(
3548
3548
  self,
transformers/models/qwen3_omni_moe/modular_qwen3_omni_moe.py CHANGED
@@ -1507,7 +1507,7 @@ class Qwen3OmniMoeTalkerCodePredictorModel(Qwen3Model):
1507
1507
  def get_input_embeddings(self):
1508
1508
  return self.codec_embedding
1509
1509
 
1510
- @check_model_inputs()
1510
+ @check_model_inputs
1511
1511
  @auto_docstring
1512
1512
  def forward(
1513
1513
  self,
transformers/models/qwen3_vl/modeling_qwen3_vl.py CHANGED
@@ -779,7 +779,7 @@ class Qwen3VLTextModel(Qwen3VLPreTrainedModel):
779
779
  # Initialize weights and apply final processing
780
780
  self.post_init()
781
781
 
782
- @check_model_inputs()
782
+ @check_model_inputs
783
783
  @auto_docstring
784
784
  def forward(
785
785
  self,
@@ -1104,7 +1104,7 @@ class Qwen3VLModel(Qwen3VLPreTrainedModel):
1104
1104
  return special_image_mask, special_video_mask
1105
1105
 
1106
1106
  @auto_docstring
1107
- @check_model_inputs()
1107
+ @check_model_inputs
1108
1108
  def forward(
1109
1109
  self,
1110
1110
  input_ids: torch.LongTensor = None,
@@ -1311,7 +1311,7 @@ class Qwen3VLForConditionalGeneration(Qwen3VLPreTrainedModel, GenerationMixin):
1311
1311
  def visual(self):
1312
1312
  return self.model.visual
1313
1313
 
1314
- @check_model_inputs()
1314
+ @check_model_inputs
1315
1315
  def forward(
1316
1316
  self,
1317
1317
  input_ids: torch.LongTensor = None,
transformers/models/qwen3_vl/modular_qwen3_vl.py CHANGED
@@ -749,7 +749,7 @@ class Qwen3VLTextModel(Qwen3VLPreTrainedModel, Qwen3Model):
749
749
  hidden_states[visual_pos_masks, :] = local_this
750
750
  return hidden_states
751
751
 
752
- @check_model_inputs()
752
+ @check_model_inputs
753
753
  @auto_docstring
754
754
  def forward(
755
755
  self,
@@ -1006,7 +1006,7 @@ class Qwen3VLModel(Qwen2_5_VLModel):
1006
1006
  return self.get_image_features(pixel_values_videos, video_grid_thw)
1007
1007
 
1008
1008
  @auto_docstring
1009
- @check_model_inputs()
1009
+ @check_model_inputs
1010
1010
  def forward(
1011
1011
  self,
1012
1012
  input_ids: torch.LongTensor = None,
@@ -1149,7 +1149,7 @@ class Qwen3VLForConditionalGeneration(Qwen2_5_VLForConditionalGeneration):
1149
1149
  config: Qwen3VLConfig
1150
1150
  _checkpoint_conversion_mapping = {}
1151
1151
 
1152
- @check_model_inputs()
1152
+ @check_model_inputs
1153
1153
  def forward(
1154
1154
  self,
1155
1155
  input_ids: torch.LongTensor = None,
transformers/models/qwen3_vl_moe/modeling_qwen3_vl_moe.py CHANGED
@@ -891,7 +891,7 @@ class Qwen3VLMoeTextModel(Qwen3VLMoePreTrainedModel):
891
891
  # Initialize weights and apply final processing
892
892
  self.post_init()
893
893
 
894
- @check_model_inputs()
894
+ @check_model_inputs
895
895
  @auto_docstring
896
896
  def forward(
897
897
  self,
@@ -1270,7 +1270,7 @@ class Qwen3VLMoeModel(Qwen3VLMoePreTrainedModel):
1270
1270
  return special_image_mask, special_video_mask
1271
1271
 
1272
1272
  @auto_docstring
1273
- @check_model_inputs()
1273
+ @check_model_inputs
1274
1274
  def forward(
1275
1275
  self,
1276
1276
  input_ids: torch.LongTensor = None,
@@ -1530,7 +1530,7 @@ class Qwen3VLMoeForConditionalGeneration(Qwen3VLMoePreTrainedModel, GenerationMi
1530
1530
  def visual(self):
1531
1531
  return self.model.visual
1532
1532
 
1533
- @check_model_inputs()
1533
+ @check_model_inputs
1534
1534
  def forward(
1535
1535
  self,
1536
1536
  input_ids: torch.LongTensor = None,
transformers/models/sam/modeling_sam.py CHANGED
@@ -1197,7 +1197,7 @@ class SamModel(SamPreTrainedModel):
1197
1197
  )
1198
1198
  return prompt_output
1199
1199
 
1200
- @check_model_inputs()
1200
+ @check_model_inputs
1201
1201
  @auto_docstring
1202
1202
  def forward(
1203
1203
  self,
transformers/models/sam2/modeling_sam2.py CHANGED
@@ -618,7 +618,7 @@ class Sam2HieraDetModel(Sam2PreTrainedModel):
618
618
  pos_embed = pos_embed.permute(0, 2, 3, 1)
619
619
  return pos_embed
620
620
 
621
- @check_model_inputs()
621
+ @check_model_inputs
622
622
  def forward(
623
623
  self,
624
624
  pixel_values: Optional[torch.FloatTensor] = None,
@@ -670,7 +670,7 @@ class Sam2VisionModel(Sam2PreTrainedModel):
670
670
  def get_input_embeddings(self):
671
671
  return self.backbone.get_input_embeddings()
672
672
 
673
- @check_model_inputs()
673
+ @check_model_inputs
674
674
  def forward(
675
675
  self,
676
676
  pixel_values: Optional[torch.FloatTensor] = None,
@@ -1387,7 +1387,7 @@ class Sam2Model(Sam2PreTrainedModel):
1387
1387
  )
1388
1388
  return prompt_output
1389
1389
 
1390
- @check_model_inputs()
1390
+ @check_model_inputs
1391
1391
  @auto_docstring
1392
1392
  def forward(
1393
1393
  self,
transformers/models/sam2/modular_sam2.py CHANGED
@@ -726,7 +726,7 @@ class Sam2HieraDetModel(Sam2PreTrainedModel):
726
726
  pos_embed = pos_embed.permute(0, 2, 3, 1)
727
727
  return pos_embed
728
728
 
729
- @check_model_inputs()
729
+ @check_model_inputs
730
730
  def forward(
731
731
  self,
732
732
  pixel_values: Optional[torch.FloatTensor] = None,
@@ -778,7 +778,7 @@ class Sam2VisionModel(Sam2PreTrainedModel):
778
778
  def get_input_embeddings(self):
779
779
  return self.backbone.get_input_embeddings()
780
780
 
781
- @check_model_inputs()
781
+ @check_model_inputs
782
782
  def forward(
783
783
  self,
784
784
  pixel_values: Optional[torch.FloatTensor] = None,
@@ -1280,7 +1280,7 @@ class Sam2Model(SamModel):
1280
1280
 
1281
1281
  return feature_maps, feature_maps_position_embeddings, vision_outputs.hidden_states, vision_outputs.attentions
1282
1282
 
1283
- @check_model_inputs()
1283
+ @check_model_inputs
1284
1284
  @auto_docstring
1285
1285
  def forward(
1286
1286
  self,
transformers/models/sam_hq/modeling_sam_hq.py CHANGED
@@ -1320,7 +1320,7 @@ class SamHQModel(SamHQPreTrainedModel):
1320
1320
  )
1321
1321
  return prompt_output
1322
1322
 
1323
- @check_model_inputs()
1323
+ @check_model_inputs
1324
1324
  @auto_docstring
1325
1325
  def forward(
1326
1326
  self,
transformers/models/seed_oss/modeling_seed_oss.py CHANGED
@@ -349,7 +349,7 @@ class SeedOssModel(SeedOssPreTrainedModel):
349
349
  # Initialize weights and apply final processing
350
350
  self.post_init()
351
351
 
352
- @check_model_inputs()
352
+ @check_model_inputs
353
353
  @auto_docstring
354
354
  def forward(
355
355
  self,
transformers/models/siglip/modeling_siglip.py CHANGED
@@ -1023,7 +1023,7 @@ class SiglipForImageClassification(SiglipPreTrainedModel):
1023
1023
  # Initialize weights and apply final processing
1024
1024
  self.post_init()
1025
1025
 
1026
- @check_model_inputs()
1026
+ @check_model_inputs
1027
1027
  @auto_docstring
1028
1028
  def forward(
1029
1029
  self,
transformers/models/siglip2/modeling_siglip2.py CHANGED
@@ -1102,7 +1102,7 @@ class Siglip2ForImageClassification(Siglip2PreTrainedModel):
1102
1102
  # Initialize weights and apply final processing
1103
1103
  self.post_init()
1104
1104
 
1105
- @check_model_inputs()
1105
+ @check_model_inputs
1106
1106
  @auto_docstring
1107
1107
  def forward(
1108
1108
  self,
transformers/models/smollm3/modeling_smollm3.py CHANGED
@@ -355,7 +355,7 @@ class SmolLM3Model(SmolLM3PreTrainedModel):
355
355
  # Initialize weights and apply final processing
356
356
  self.post_init()
357
357
 
358
- @check_model_inputs()
358
+ @check_model_inputs
359
359
  @auto_docstring
360
360
  def forward(
361
361
  self,
transformers/models/starcoder2/modeling_starcoder2.py CHANGED
@@ -325,7 +325,7 @@ class Starcoder2Model(Starcoder2PreTrainedModel):
325
325
  # Initialize weights and apply final processing
326
326
  self.post_init()
327
327
 
328
- @check_model_inputs()
328
+ @check_model_inputs
329
329
  def forward(
330
330
  self,
331
331
  input_ids: Optional[torch.LongTensor] = None,