[GH-ISSUE #10] 'Qwen3TTSTalkerConfig' object has no attribute 'pad_token_id' #9

New issue

Open

opened 2026-10-30 21:03:22 +00:00 by bsakaguchi · 4 comments

bsakaguchi commented 2026-10-30 21:03:22 +00:00

Owner

Copy link

Originally created by @DNPMBHC on GitHub (Jan 27, 2026).
Original GitHub issue: https://github.com/1038lab/ComfyUI-QwenTTS/issues/10

ComfyUI Error Report

Error Details

• Node ID: 45
• Node Type: AILab_Qwen3TTSCustomVoice
• Exception Type: AttributeError
• Exception Message: 'Qwen3TTSTalkerConfig' object has no attribute 'pad_token_id'

Stack Trace

  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\execution.py", line 518, in execute
    output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data)
                                                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\execution.py", line 329, in get_output_data
    return_values = await _async_map_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\execution.py", line 303, in _async_map_node_over_list
    await process_inputs(input_dict, i)

  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\execution.py", line 291, in process_inputs
    result = f(**inputs)
             ^^^^^^^^^^^

  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\AILab_QwenTTS.py", line 554, in generate
    return _custom_voice_generate(
           ^^^^^^^^^^^^^^^^^^^^^^^

  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\AILab_QwenTTS.py", line 419, in _custom_voice_generate
    model = _load_model("CustomVoice", model_size, device, precision)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\AILab_QwenTTS.py", line 317, in _load_model
    model = Qwen3TTSModel.from_pretrained(source, device_map=device, dtype=dtype)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\inference\qwen3_tts_model.py", line 112, in from_pretrained
    model = AutoModel.from_pretrained(pretrained_model_name_or_path, **kwargs)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\models\auto\auto_factory.py", line 372, in from_pretrained
    return model_class.from_pretrained(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1872, in from_pretrained
    model = super().from_pretrained(
            ^^^^^^^^^^^^^^^^^^^^^^^^

  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 4072, in from_pretrained
    model = cls(config, *model_args, **model_kwargs)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1821, in __init__
    self.talker = Qwen3TTSTalkerForConditionalGeneration(self.config.talker_config)
                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1574, in __init__
    self.model = Qwen3TTSTalkerModel(config)
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1434, in __init__
    self.padding_idx = config.pad_token_id
                       ^^^^^^^^^^^^^^^^^^^

  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\configuration_utils.py", line 164, in __getattribute__
    return super().__getattribute__(key)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

System Information

• ComfyUI Version: 0.11.0
• Arguments: ComfyUI\main.py --windows-standalone-build
• OS: win32
• Python Version: 3.12.10 (tags/v3.12.10:0cc8128, Apr 8 2025, 12:21:36) [MSC v.1943 64 bit (AMD64)]
• Embedded Python: true
• PyTorch Version: 2.11.0.dev20260115+cu130

Devices

• Name: cuda:0 NVIDIA GeForce RTX 5060 Ti : cudaMallocAsync
  • Type: cuda
  • VRAM Total: 17102864384
  • VRAM Free: 15896412160
  • Torch VRAM Total: 0
  • Torch VRAM Free: 0

Logs

2026-01-27T18:51:29.948942 - [START] Security scan2026-01-27T18:51:29.948942 - 
2026-01-27T18:51:32.286113 - [DONE] Security scan2026-01-27T18:51:32.286113 - 
2026-01-27T18:51:32.592772 - ## ComfyUI-Manager: installing dependencies done.2026-01-27T18:51:32.592772 - 
2026-01-27T18:51:32.592772 - ** ComfyUI startup time:2026-01-27T18:51:32.592772 -  2026-01-27T18:51:32.592772 - 2026-01-27 18:51:32.5922026-01-27T18:51:32.592772 - 
2026-01-27T18:51:32.592772 - ** Platform:2026-01-27T18:51:32.593747 -  2026-01-27T18:51:32.593747 - Windows2026-01-27T18:51:32.593747 - 
2026-01-27T18:51:32.593747 - ** Python version:2026-01-27T18:51:32.593747 -  2026-01-27T18:51:32.593747 - 3.12.10 (tags/v3.12.10:0cc8128, Apr  8 2025, 12:21:36) [MSC v.1943 64 bit (AMD64)]2026-01-27T18:51:32.593747 - 
2026-01-27T18:51:32.593747 - ** Python executable:2026-01-27T18:51:32.593747 -  2026-01-27T18:51:32.593747 - E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\python.exe2026-01-27T18:51:32.593747 - 
2026-01-27T18:51:32.594747 - ** ComfyUI Path:2026-01-27T18:51:32.594747 -  2026-01-27T18:51:32.594747 - E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI2026-01-27T18:51:32.594747 - 
2026-01-27T18:51:32.594747 - ** ComfyUI Base Folder Path:2026-01-27T18:51:32.594747 -  2026-01-27T18:51:32.594747 - E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI2026-01-27T18:51:32.594747 - 
2026-01-27T18:51:32.594747 - ** User directory:2026-01-27T18:51:32.594747 -  2026-01-27T18:51:32.594747 - E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\user2026-01-27T18:51:32.602906 - 
2026-01-27T18:51:32.604000 - ** ComfyUI-Manager config path:2026-01-27T18:51:32.604000 -  2026-01-27T18:51:32.604000 - E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\user\__manager\config.ini2026-01-27T18:51:32.604000 - 
2026-01-27T18:51:32.604000 - ** Log path:2026-01-27T18:51:32.604000 -  2026-01-27T18:51:32.604000 - E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\user\comfyui.log2026-01-27T18:51:32.604000 - 
2026-01-27T18:51:35.026670 - 
Prestartup times for custom nodes:
2026-01-27T18:51:35.026670 -    5.6 seconds: E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-manager
2026-01-27T18:51:35.026670 - 
2026-01-27T18:51:35.760805 - E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\cuda\__init__.py:61: FutureWarning: The pynvml package is deprecated. Please install nvidia-ml-py instead. If you did not install pynvml directly, please report this to the maintainers of the package that installed pynvml for you.
  import pynvml  # type: ignore[import]
2026-01-27T18:51:38.886250 - Checkpoint files will always be loaded safely.
2026-01-27T18:51:39.029780 - Total VRAM 16311 MB, total RAM 65434 MB
2026-01-27T18:51:39.029780 - pytorch version: 2.11.0.dev20260115+cu130
2026-01-27T18:51:39.030778 - Set vram state to: NORMAL_VRAM
2026-01-27T18:51:39.031780 - Device: cuda:0 NVIDIA GeForce RTX 5060 Ti : cudaMallocAsync
2026-01-27T18:51:39.052881 - Using async weight offloading with 2 streams
2026-01-27T18:51:39.054386 - Enabled pinned memory 29445.0
2026-01-27T18:51:41.527823 - Found comfy_kitchen backend cuda: {'available': True, 'disabled': False, 'unavailable_reason': None, 'capabilities': ['apply_rope', 'apply_rope1', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8', 'scaled_mm_nvfp4']}
2026-01-27T18:51:41.528861 - Found comfy_kitchen backend eager: {'available': True, 'disabled': False, 'unavailable_reason': None, 'capabilities': ['apply_rope', 'apply_rope1', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8', 'scaled_mm_nvfp4']}
2026-01-27T18:51:41.528861 - Found comfy_kitchen backend triton: {'available': True, 'disabled': True, 'unavailable_reason': None, 'capabilities': ['apply_rope', 'apply_rope1', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8']}
2026-01-27T18:51:42.156347 - Using pytorch attention
2026-01-27T18:51:48.382001 - Python version: 3.12.10 (tags/v3.12.10:0cc8128, Apr  8 2025, 12:21:36) [MSC v.1943 64 bit (AMD64)]
2026-01-27T18:51:48.382001 - ComfyUI version: 0.11.0
2026-01-27T18:51:48.508878 - ComfyUI frontend version: 1.37.11
2026-01-27T18:51:48.512888 - [Prompt Server] web root: E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\comfyui_frontend_package\static
2026-01-27T18:51:50.585924 - [Crystools INFO] Crystools version: 1.27.4
2026-01-27T18:51:50.640074 - [Crystools INFO] Platform release: 11
2026-01-27T18:51:50.640074 - [Crystools INFO] JETSON: Not detected.
2026-01-27T18:51:50.642072 - [Crystools INFO] CPU: Intel(R) Core(TM) i3-10105 CPU @ 3.70GHz - Arch: AMD64 - OS: Windows 11
2026-01-27T18:51:50.654726 - [Crystools INFO] pynvml (NVIDIA) initialized.
2026-01-27T18:51:50.655728 - [Crystools INFO] GPU/s:
2026-01-27T18:51:50.673379 - [Crystools INFO] 0) NVIDIA GeForce RTX 5060 Ti
2026-01-27T18:51:50.673379 - [Crystools INFO] NVIDIA Driver: 591.74
2026-01-27T18:51:50.696692 - ### Loading: ComfyUI-Manager (V3.39.2)
2026-01-27T18:51:50.697702 - [ComfyUI-Manager] network_mode: public
2026-01-27T18:51:50.697702 - [ComfyUI-Manager] ComfyUI per-queue preview override detected (PR #11261). Manager's preview method feature is disabled. Use ComfyUI's --preview-method CLI option or 'Settings > Execution > Live preview method'.
2026-01-27T18:51:51.130856 - ### ComfyUI Version: v0.11.0 | Released on '2026-01-26'
2026-01-27T18:51:51.836698 - Warning: Unable to access E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\models\diffusion_models\wan2.2_i2v_14B_fp8_scaled. Skipping this path.
2026-01-27T18:51:51.837708 - Warning: Unable to access E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\models\diffusion_models\Wan2_2-T2V-A14B fp8. Skipping this path.
2026-01-27T18:51:51.875093 - 
Import times for custom nodes:
2026-01-27T18:51:51.876093 -    0.0 seconds: E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\websocket_image_save.py
2026-01-27T18:51:51.876093 -    0.0 seconds: E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-ReservedVRAM-main
2026-01-27T18:51:51.876093 -    0.0 seconds: E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-SageAttention3
2026-01-27T18:51:51.876093 -    0.0 seconds: E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS
2026-01-27T18:51:51.876093 -    0.0 seconds: E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-various-main
2026-01-27T18:51:51.877093 -    0.1 seconds: E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Crystools-main
2026-01-27T18:51:51.877093 -    0.5 seconds: E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-manager
2026-01-27T18:51:51.877093 -    0.7 seconds: E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_Swwan
2026-01-27T18:51:51.877093 - 
2026-01-27T18:51:51.886142 - Context impl SQLiteImpl.
2026-01-27T18:51:51.886142 - Will assume non-transactional DDL.
2026-01-27T18:51:52.079533 - Assets scan(roots=['models']) completed in 0.185s (created=0, skipped_existing=111, total_seen=122)
2026-01-27T18:51:52.127119 - [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/alter-list.json
2026-01-27T18:51:52.300139 - Starting server

2026-01-27T18:51:52.301107 - To see the GUI go to: http://127.0.0.1:8188
2026-01-27T18:51:52.337612 - [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/model-list.json
2026-01-27T18:51:52.526812 - [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/github-stats.json
2026-01-27T18:51:52.944291 - [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/extension-node-map.json
2026-01-27T18:51:53.527647 - [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/custom-node-list.json
2026-01-27T18:51:53.770576 - [DEPRECATION WARNING] Detected import of deprecated legacy API: /scripts/ui/components/buttonGroup.js. This is likely caused by a custom node extension using outdated APIs. Please update your extensions or contact the extension author for an updated version.
2026-01-27T18:51:53.772592 - [DEPRECATION WARNING] Detected import of deprecated legacy API: /scripts/ui.js. This is likely caused by a custom node extension using outdated APIs. Please update your extensions or contact the extension author for an updated version.
2026-01-27T18:51:53.773592 - [DEPRECATION WARNING] Detected import of deprecated legacy API: /extensions/core/groupNode.js. This is likely caused by a custom node extension using outdated APIs. Please update your extensions or contact the extension author for an updated version.
2026-01-27T18:51:54.861922 - [DEPRECATION WARNING] Detected import of deprecated legacy API: /scripts/ui/components/button.js. This is likely caused by a custom node extension using outdated APIs. Please update your extensions or contact the extension author for an updated version.
2026-01-27T18:51:58.164217 - got prompt
2026-01-27T18:51:59.586616 - speaker_encoder_config is None. Initializing talker model with default values
2026-01-27T18:51:59.587654 - E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_rope_utils.py:927: FutureWarning: `rope_config_validation` is deprecated and has been removed. Its functionality has been moved to RotaryEmbeddingConfigMixin.validate_rope method. PreTrainedConfig inherits this class, so please call self.validate_rope() instead. Also, make sure to use the new rope_parameters syntax. You can call self.standardize_rope_params() in the meantime.
  warnings.warn(
2026-01-27T18:51:59.589161 - talker_config is None. Initializing talker model with default values
2026-01-27T18:51:59.589161 - speaker_encoder_config is None. Initializing talker model with default values
2026-01-27T18:51:59.590171 - code_predictor_config is None. Initializing code_predictor model with default values
2026-01-27T18:51:59.591169 - code_predictor_config is None. Initializing code_predictor model with default values
2026-01-27T18:51:59.595166 - [Qwen3-TTS] flash_attention_2 unavailable, fallback to default attention2026-01-27T18:51:59.595166 - 
2026-01-27T18:51:59.598682 - speaker_encoder_config is None. Initializing talker model with default values
2026-01-27T18:51:59.599680 - talker_config is None. Initializing talker model with default values
2026-01-27T18:51:59.599680 - speaker_encoder_config is None. Initializing talker model with default values
2026-01-27T18:51:59.599680 - code_predictor_config is None. Initializing code_predictor model with default values
2026-01-27T18:51:59.600679 - code_predictor_config is None. Initializing code_predictor model with default values
2026-01-27T18:51:59.607833 - !!! Exception during processing !!! 'Qwen3TTSTalkerConfig' object has no attribute 'pad_token_id'
2026-01-27T18:51:59.621879 - Traceback (most recent call last):
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\AILab_QwenTTS.py", line 311, in _load_model
    model = Qwen3TTSModel.from_pretrained(source, **kwargs)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\inference\qwen3_tts_model.py", line 112, in from_pretrained
    model = AutoModel.from_pretrained(pretrained_model_name_or_path, **kwargs)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\models\auto\auto_factory.py", line 372, in from_pretrained
    return model_class.from_pretrained(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1872, in from_pretrained
    model = super().from_pretrained(
            ^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 4072, in from_pretrained
    model = cls(config, *model_args, **model_kwargs)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1818, in __init__
    super().__init__(config)
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 1295, in __init__
    self.config._attn_implementation_internal = self._check_and_adjust_attn_implementation(
                                                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 1873, in _check_and_adjust_attn_implementation
    applicable_attn_implementation = self.get_correct_attn_implementation(
                                     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 1914, in get_correct_attn_implementation
    self._flash_attn_2_can_dispatch(is_init_check)
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 1594, in _flash_attn_2_can_dispatch
    raise ImportError(f"{preface} the package flash_attn seems to be not installed. {install_message}")
ImportError: FlashAttention2 has been toggled on, but it cannot be used due to the following error: the package flash_attn seems to be not installed. Please refer to the documentation of https://huggingface.co/docs/transformers/perf_infer_gpu_one#flashattention-2 to install Flash Attention 2.

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\execution.py", line 518, in execute
    output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data)
                                                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\execution.py", line 329, in get_output_data
    return_values = await _async_map_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\execution.py", line 303, in _async_map_node_over_list
    await process_inputs(input_dict, i)
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\execution.py", line 291, in process_inputs
    result = f(**inputs)
             ^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\AILab_QwenTTS.py", line 644, in generate
    return _voice_design_generate(
           ^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\AILab_QwenTTS.py", line 460, in _voice_design_generate
    model = _load_model("VoiceDesign", model_size, device, precision)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\AILab_QwenTTS.py", line 317, in _load_model
    model = Qwen3TTSModel.from_pretrained(source, device_map=device, dtype=dtype)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\inference\qwen3_tts_model.py", line 112, in from_pretrained
    model = AutoModel.from_pretrained(pretrained_model_name_or_path, **kwargs)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\models\auto\auto_factory.py", line 372, in from_pretrained
    return model_class.from_pretrained(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1872, in from_pretrained
    model = super().from_pretrained(
            ^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 4072, in from_pretrained
    model = cls(config, *model_args, **model_kwargs)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1821, in __init__
    self.talker = Qwen3TTSTalkerForConditionalGeneration(self.config.talker_config)
                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1574, in __init__
    self.model = Qwen3TTSTalkerModel(config)
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1434, in __init__
    self.padding_idx = config.pad_token_id
                       ^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\configuration_utils.py", line 164, in __getattribute__
    return super().__getattribute__(key)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
AttributeError: 'Qwen3TTSTalkerConfig' object has no attribute 'pad_token_id'

2026-01-27T18:51:59.627399 - Prompt executed in 1.46 seconds
2026-01-27T18:52:02.534882 - FETCH ComfyRegistry Data: 5/1222026-01-27T18:52:02.534882 - 
2026-01-27T18:52:11.762719 - FETCH ComfyRegistry Data: 10/1222026-01-27T18:52:11.763386 - 
2026-01-27T18:52:20.401999 - FETCH ComfyRegistry Data: 15/1222026-01-27T18:52:20.401999 - 
2026-01-27T18:52:28.744270 - FETCH ComfyRegistry Data: 20/1222026-01-27T18:52:28.745269 - 
2026-01-27T18:52:36.897609 - FETCH ComfyRegistry Data: 25/1222026-01-27T18:52:36.897609 - 
2026-01-27T18:52:46.058172 - FETCH ComfyRegistry Data: 30/1222026-01-27T18:52:46.058676 - 
2026-01-27T18:52:55.796253 - FETCH ComfyRegistry Data: 35/1222026-01-27T18:52:55.796253 - 
2026-01-27T18:53:03.980159 - FETCH ComfyRegistry Data: 40/1222026-01-27T18:53:03.982185 - 
2026-01-27T18:53:12.278949 - FETCH ComfyRegistry Data: 45/1222026-01-27T18:53:12.278949 - 
2026-01-27T18:53:21.079325 - FETCH ComfyRegistry Data: 50/1222026-01-27T18:53:21.079998 - 
2026-01-27T18:53:30.454535 - FETCH ComfyRegistry Data: 55/1222026-01-27T18:53:30.454535 - 
2026-01-27T18:53:38.713102 - FETCH ComfyRegistry Data: 60/1222026-01-27T18:53:38.714131 - 
2026-01-27T18:53:48.433411 - FETCH ComfyRegistry Data: 65/1222026-01-27T18:53:48.433411 - 
2026-01-27T18:53:58.045134 - FETCH ComfyRegistry Data: 70/1222026-01-27T18:53:58.045134 - 
2026-01-27T18:54:08.422011 - FETCH ComfyRegistry Data: 75/1222026-01-27T18:54:08.422524 - 
2026-01-27T18:54:17.137583 - FETCH ComfyRegistry Data: 80/1222026-01-27T18:54:17.137583 - 
2026-01-27T18:54:25.472192 - FETCH ComfyRegistry Data: 85/1222026-01-27T18:54:25.472192 - 
2026-01-27T18:54:34.968890 - FETCH ComfyRegistry Data: 90/1222026-01-27T18:54:34.968890 - 
2026-01-27T18:54:43.334997 - FETCH ComfyRegistry Data: 95/1222026-01-27T18:54:43.334997 - 
2026-01-27T18:54:56.398747 - FETCH ComfyRegistry Data: 100/1222026-01-27T18:54:56.398747 - 
2026-01-27T18:55:05.589262 - FETCH ComfyRegistry Data: 105/1222026-01-27T18:55:05.589262 - 
2026-01-27T18:55:14.273031 - FETCH ComfyRegistry Data: 110/1222026-01-27T18:55:14.273031 - 
2026-01-27T18:55:22.980476 - FETCH ComfyRegistry Data: 115/1222026-01-27T18:55:22.980476 - 
2026-01-27T18:55:31.563040 - FETCH ComfyRegistry Data: 120/1222026-01-27T18:55:31.563040 - 
2026-01-27T18:55:35.089320 - FETCH ComfyRegistry Data [DONE]2026-01-27T18:55:35.091349 - 
2026-01-27T18:55:35.363256 - [ComfyUI-Manager] default cache updated: https://api.comfy.org/nodes
2026-01-27T18:55:35.400501 - FETCH DATA from: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/custom-node-list.json2026-01-27T18:55:37.166808 -  [DONE]2026-01-27T18:55:37.167834 - 
2026-01-27T18:55:37.405881 - [ComfyUI-Manager] All startup tasks have been completed.
2026-01-27T18:57:24.247018 - got prompt
2026-01-27T18:57:24.268084 - speaker_encoder_config is None. Initializing talker model with default values
2026-01-27T18:57:24.269595 - talker_config is None. Initializing talker model with default values
2026-01-27T18:57:24.270607 - speaker_encoder_config is None. Initializing talker model with default values
2026-01-27T18:57:24.272632 - code_predictor_config is None. Initializing code_predictor model with default values
2026-01-27T18:57:24.274626 - code_predictor_config is None. Initializing code_predictor model with default values
2026-01-27T18:57:24.286181 - [Qwen3-TTS] flash_attention_2 unavailable, fallback to default attention2026-01-27T18:57:24.287194 - 
2026-01-27T18:57:24.293098 - speaker_encoder_config is None. Initializing talker model with default values
2026-01-27T18:57:24.294875 - talker_config is None. Initializing talker model with default values
2026-01-27T18:57:24.294875 - speaker_encoder_config is None. Initializing talker model with default values
2026-01-27T18:57:24.295421 - code_predictor_config is None. Initializing code_predictor model with default values
2026-01-27T18:57:24.295963 - code_predictor_config is None. Initializing code_predictor model with default values
2026-01-27T18:57:24.307589 - !!! Exception during processing !!! 'Qwen3TTSTalkerConfig' object has no attribute 'pad_token_id'
2026-01-27T18:57:24.314627 - Traceback (most recent call last):
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\AILab_QwenTTS.py", line 311, in _load_model
    model = Qwen3TTSModel.from_pretrained(source, **kwargs)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\inference\qwen3_tts_model.py", line 112, in from_pretrained
    model = AutoModel.from_pretrained(pretrained_model_name_or_path, **kwargs)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\models\auto\auto_factory.py", line 372, in from_pretrained
    return model_class.from_pretrained(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1872, in from_pretrained
    model = super().from_pretrained(
            ^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 4072, in from_pretrained
    model = cls(config, *model_args, **model_kwargs)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1818, in __init__
    super().__init__(config)
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 1295, in __init__
    self.config._attn_implementation_internal = self._check_and_adjust_attn_implementation(
                                                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 1873, in _check_and_adjust_attn_implementation
    applicable_attn_implementation = self.get_correct_attn_implementation(
                                     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 1914, in get_correct_attn_implementation
    self._flash_attn_2_can_dispatch(is_init_check)
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 1594, in _flash_attn_2_can_dispatch
    raise ImportError(f"{preface} the package flash_attn seems to be not installed. {install_message}")
ImportError: FlashAttention2 has been toggled on, but it cannot be used due to the following error: the package flash_attn seems to be not installed. Please refer to the documentation of https://huggingface.co/docs/transformers/perf_infer_gpu_one#flashattention-2 to install Flash Attention 2.

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\execution.py", line 518, in execute
    output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data)
                                                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\execution.py", line 329, in get_output_data
    return_values = await _async_map_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\execution.py", line 303, in _async_map_node_over_list
    await process_inputs(input_dict, i)
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\execution.py", line 291, in process_inputs
    result = f(**inputs)
             ^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\AILab_QwenTTS.py", line 554, in generate
    return _custom_voice_generate(
           ^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\AILab_QwenTTS.py", line 419, in _custom_voice_generate
    model = _load_model("CustomVoice", model_size, device, precision)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\AILab_QwenTTS.py", line 317, in _load_model
    model = Qwen3TTSModel.from_pretrained(source, device_map=device, dtype=dtype)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\inference\qwen3_tts_model.py", line 112, in from_pretrained
    model = AutoModel.from_pretrained(pretrained_model_name_or_path, **kwargs)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\models\auto\auto_factory.py", line 372, in from_pretrained
    return model_class.from_pretrained(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1872, in from_pretrained
    model = super().from_pretrained(
            ^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 4072, in from_pretrained
    model = cls(config, *model_args, **model_kwargs)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1821, in __init__
    self.talker = Qwen3TTSTalkerForConditionalGeneration(self.config.talker_config)
                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1574, in __init__
    self.model = Qwen3TTSTalkerModel(config)
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1434, in __init__
    self.padding_idx = config.pad_token_id
                       ^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\configuration_utils.py", line 164, in __getattribute__
    return super().__getattribute__(key)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
AttributeError: 'Qwen3TTSTalkerConfig' object has no attribute 'pad_token_id'

2026-01-27T18:57:24.320144 - Prompt executed in 0.07 seconds
2026-01-27T18:58:29.028882 - got prompt
2026-01-27T18:58:29.046458 - speaker_encoder_config is None. Initializing talker model with default values
2026-01-27T18:58:29.048771 - talker_config is None. Initializing talker model with default values
2026-01-27T18:58:29.048771 - speaker_encoder_config is None. Initializing talker model with default values
2026-01-27T18:58:29.048771 - code_predictor_config is None. Initializing code_predictor model with default values
2026-01-27T18:58:29.049796 - code_predictor_config is None. Initializing code_predictor model with default values
2026-01-27T18:58:29.054863 - [Qwen3-TTS] flash_attention_2 unavailable, fallback to default attention2026-01-27T18:58:29.054863 - 
2026-01-27T18:58:29.064403 - speaker_encoder_config is None. Initializing talker model with default values
2026-01-27T18:58:29.068926 - talker_config is None. Initializing talker model with default values
2026-01-27T18:58:29.069937 - speaker_encoder_config is None. Initializing talker model with default values
2026-01-27T18:58:29.071946 - code_predictor_config is None. Initializing code_predictor model with default values
2026-01-27T18:58:29.072948 - code_predictor_config is None. Initializing code_predictor model with default values
2026-01-27T18:58:29.085010 - !!! Exception during processing !!! 'Qwen3TTSTalkerConfig' object has no attribute 'pad_token_id'
2026-01-27T18:58:29.094118 - Traceback (most recent call last):
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\AILab_QwenTTS.py", line 311, in _load_model
    model = Qwen3TTSModel.from_pretrained(source, **kwargs)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\inference\qwen3_tts_model.py", line 112, in from_pretrained
    model = AutoModel.from_pretrained(pretrained_model_name_or_path, **kwargs)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\models\auto\auto_factory.py", line 372, in from_pretrained
    return model_class.from_pretrained(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1872, in from_pretrained
    model = super().from_pretrained(
            ^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 4072, in from_pretrained
    model = cls(config, *model_args, **model_kwargs)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1818, in __init__
    super().__init__(config)
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 1295, in __init__
    self.config._attn_implementation_internal = self._check_and_adjust_attn_implementation(
                                                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 1873, in _check_and_adjust_attn_implementation
    applicable_attn_implementation = self.get_correct_attn_implementation(
                                     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 1914, in get_correct_attn_implementation
    self._flash_attn_2_can_dispatch(is_init_check)
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 1594, in _flash_attn_2_can_dispatch
    raise ImportError(f"{preface} the package flash_attn seems to be not installed. {install_message}")
ImportError: FlashAttention2 has been toggled on, but it cannot be used due to the following error: the package flash_attn seems to be not installed. Please refer to the documentation of https://huggingface.co/docs/transformers/perf_infer_gpu_one#flashattention-2 to install Flash Attention 2.

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\execution.py", line 518, in execute
    output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data)
                                                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\execution.py", line 329, in get_output_data
    return_values = await _async_map_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\execution.py", line 303, in _async_map_node_over_list
    await process_inputs(input_dict, i)
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\execution.py", line 291, in process_inputs
    result = f(**inputs)
             ^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\AILab_QwenTTS.py", line 554, in generate
    return _custom_voice_generate(
           ^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\AILab_QwenTTS.py", line 419, in _custom_voice_generate
    model = _load_model("CustomVoice", model_size, device, precision)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\AILab_QwenTTS.py", line 317, in _load_model
    model = Qwen3TTSModel.from_pretrained(source, device_map=device, dtype=dtype)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\inference\qwen3_tts_model.py", line 112, in from_pretrained
    model = AutoModel.from_pretrained(pretrained_model_name_or_path, **kwargs)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\models\auto\auto_factory.py", line 372, in from_pretrained
    return model_class.from_pretrained(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1872, in from_pretrained
    model = super().from_pretrained(
            ^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 4072, in from_pretrained
    model = cls(config, *model_args, **model_kwargs)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1821, in __init__
    self.talker = Qwen3TTSTalkerForConditionalGeneration(self.config.talker_config)
                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1574, in __init__
    self.model = Qwen3TTSTalkerModel(config)
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1434, in __init__
    self.padding_idx = config.pad_token_id
                       ^^^^^^^^^^^^^^^^^^^
  File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\configuration_utils.py", line 164, in __getattribute__
    return super().__getattribute__(key)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
AttributeError: 'Qwen3TTSTalkerConfig' object has no attribute 'pad_token_id'

2026-01-27T18:58:29.098395 - Prompt executed in 0.07 seconds

Attached Workflow

Please make sure that workflow does not contain any sensitive information such as API keys or passwords.

{"id":"6b2e5d09-d1ea-4fc9-a8bd-8daca49d65ff","revision":0,"last_node_id":52,"last_link_id":38,"nodes":[{"id":46,"type":"AILab_Qwen3TTSVoiceClone","pos":[2.1580014941039964,947.4393213149233],"size":[340,262],"flags":{},"order":4,"mode":4,"inputs":[{"localized_name":"reference_audio","name":"reference_audio","type":"AUDIO","link":35},{"localized_name":"target_text","name":"target_text","type":"STRING","widget":{"name":"target_text"},"link":null},{"localized_name":"model_size","name":"model_size","type":"COMBO","widget":{"name":"model_size"},"link":null},{"localized_name":"language","name":"language","type":"COMBO","widget":{"name":"language"},"link":null},{"localized_name":"reference_text","name":"reference_text","shape":7,"type":"STRING","widget":{"name":"reference_text"},"link":null},{"localized_name":"x_vector_only","name":"x_vector_only","shape":7,"type":"BOOLEAN","widget":{"name":"x_vector_only"},"link":null},{"localized_name":"seed","name":"seed","shape":7,"type":"INT","widget":{"name":"seed"},"link":null}],"outputs":[{"localized_name":"音频","name":"AUDIO","type":"AUDIO","links":[37]}],"properties":{"cnr_id":"comfyui-qwentts","ver":"1.0.1","Node name for S&R":"AILab_Qwen3TTSVoiceClone","ue_properties":{"widget_ue_connectable":{},"input_ue_unconnectable":{},"version":"7.5.2"}},"widgets_values":["Hello, this is a cloned voice.","0.6B","Auto","",false,875958189094423,"randomize"],"color":"#4D2600","bgcolor":"#3e5249"},{"id":47,"type":"LoadAudio","pos":[2.1580014941039964,767.8369715781096],"size":[340,136],"flags":{},"order":0,"mode":4,"inputs":[{"localized_name":"音频","name":"audio","type":"COMBO","widget":{"name":"audio"},"link":null},{"localized_name":"音频UI","name":"audioUI","type":"AUDIO_UI","widget":{"name":"audioUI"},"link":null},{"localized_name":"选择文件上传","name":"upload","type":"AUDIOUPLOAD","widget":{"name":"upload"},"link":null}],"outputs":[{"localized_name":"音频","name":"AUDIO","type":"AUDIO","links":[35]}],"properties":{"cnr_id":"comfy-core","ver":"0.10.0","Node name for S&R":"LoadAudio","ue_properties":{"widget_ue_connectable":{},"input_ue_unconnectable":{}}},"widgets_values":["107289363.mp4",null,null]},{"id":51,"type":"SaveAudioMP3","pos":[2.1580014941039964,1260.041137629123],"size":[340,136],"flags":{},"order":7,"mode":4,"inputs":[{"localized_name":"音频","name":"audio","type":"AUDIO","link":37},{"localized_name":"文件名前缀","name":"filename_prefix","type":"STRING","widget":{"name":"filename_prefix"},"link":null},{"localized_name":"质量","name":"quality","type":"COMBO","widget":{"name":"quality"},"link":null},{"localized_name":"audioUI","name":"audioUI","type":"AUDIO_UI","widget":{"name":"audioUI"},"link":null}],"outputs":[],"properties":{"cnr_id":"comfy-core","ver":"0.10.0","Node name for S&R":"SaveAudioMP3","ue_properties":{"widget_ue_connectable":{},"input_ue_unconnectable":{}}},"widgets_values":["audio/ComfyUI","320k"]},{"id":49,"type":"SaveAudioMP3","pos":[368.7138198400269,1260.041137629123],"size":[340,136],"flags":{},"order":5,"mode":0,"inputs":[{"localized_name":"音频","name":"audio","type":"AUDIO","link":38},{"localized_name":"文件名前缀","name":"filename_prefix","type":"STRING","widget":{"name":"filename_prefix"},"link":null},{"localized_name":"质量","name":"quality","type":"COMBO","widget":{"name":"quality"},"link":null},{"localized_name":"audioUI","name":"audioUI","type":"AUDIO_UI","widget":{"name":"audioUI"},"link":null}],"outputs":[],"properties":{"cnr_id":"comfy-core","ver":"0.10.0","Node name for S&R":"SaveAudioMP3","ue_properties":{"widget_ue_connectable":{},"input_ue_unconnectable":{}}},"widgets_values":["audio/ComfyUI","V0"]},{"id":45,"type":"AILab_Qwen3TTSCustomVoice","pos":[368.7138198400269,947.4393213149233],"size":[340,262],"flags":{},"order":1,"mode":0,"inputs":[{"localized_name":"text","name":"text","type":"STRING","widget":{"name":"text"},"link":null},{"localized_name":"speaker","name":"speaker","type":"COMBO","widget":{"name":"speaker"},"link":null},{"localized_name":"model_size","name":"model_size","type":"COMBO","widget":{"name":"model_size"},"link":null},{"localized_name":"language","name":"language","type":"COMBO","widget":{"name":"language"},"link":null},{"localized_name":"instruct","name":"instruct","shape":7,"type":"STRING","widget":{"name":"instruct"},"link":null},{"localized_name":"seed","name":"seed","shape":7,"type":"INT","widget":{"name":"seed"},"link":null}],"outputs":[{"localized_name":"音频","name":"AUDIO","type":"AUDIO","links":[38]}],"properties":{"cnr_id":"comfyui-qwentts","ver":"1.0.1","Node name for S&R":"AILab_Qwen3TTSCustomVoice","ue_properties":{"widget_ue_connectable":{},"input_ue_unconnectable":{},"version":"7.5.2"}},"widgets_values":["Hello from Qwen3-TTS.","Ryan","1.7B","Auto","",342286630204760,"randomize"],"color":"#4D2600","bgcolor":"#3e5249"},{"id":41,"type":"AILab_Qwen3TTSVoiceDesign","pos":[737.700114583396,945.6163481165029],"size":[340,238],"flags":{},"order":2,"mode":4,"inputs":[{"localized_name":"text","name":"text","type":"STRING","widget":{"name":"text"},"link":null},{"localized_name":"instruct","name":"instruct","type":"STRING","widget":{"name":"instruct"},"link":null},{"localized_name":"model_size","name":"model_size","type":"COMBO","widget":{"name":"model_size"},"link":null},{"localized_name":"language","name":"language","type":"COMBO","widget":{"name":"language"},"link":null},{"localized_name":"seed","name":"seed","shape":7,"type":"INT","widget":{"name":"seed"},"link":null}],"outputs":[{"localized_name":"音频","name":"AUDIO","type":"AUDIO","links":[36]}],"properties":{"cnr_id":"comfyui-qwentts","ver":"1.0.1","Node name for S&R":"AILab_Qwen3TTSVoiceDesign","ue_properties":{"widget_ue_connectable":{},"input_ue_unconnectable":{},"version":"7.5.2"}},"widgets_values":["Hello from Qwen3-TTS VoiceDesign.","A warm, gentle female voice.","1.7B","Auto",242260717380428,"randomize"],"color":"#4D2600","bgcolor":"#3e5249"},{"id":52,"type":"MarkdownNote","pos":[-418.20320401527084,760.3485979843861],"size":[383.75722406036607,599.0649205241322],"flags":{},"order":3,"mode":0,"inputs":[],"outputs":[],"properties":{},"widgets_values":["# ComfyUI-QwenTTS\n\nA clean, efficient ComfyUI custom node pack for **Qwen3-TTS**. It provides **CustomVoice**, **VoiceDesign**, and **VoiceClone** workflows with strict ComfyUI compatibility and practical controls for quality, speed, and stability.\n\n## Usage\n\n### Custom Voice TTS\nAdd **Qwen3 TTS CustomVoice** and fill `text`, choose `speaker`, optionally add `instruct`.\n\n### Voice Design\nAdd **Qwen3 TTS VoiceDesign**, provide `text` and `instruct`.\n\n### Voice Clone\nAdd **Qwen3 TTS VoiceClone**, connect `reference_audio`, and provide `reference_text`.\n\n## Model Download\n\nModels can be auto-downloaded to:\n```\nComfyUI/models/TTS/Qwen3-TTS/<MODEL_NAME>/\n```\n\nSupported model IDs (Hugging Face):\n- [Qwen/Qwen3-TTS-12Hz-1.7B-CustomVoice](https://huggingface.co/Qwen/Qwen3-TTS-12Hz-1.7B-CustomVoice)\n- [Qwen/Qwen3-TTS-12Hz-0.6B-CustomVoice](https://huggingface.co/Qwen/Qwen3-TTS-12Hz-0.6B-CustomVoice)\n- [Qwen/Qwen3-TTS-12Hz-1.7B-VoiceDesign](https://huggingface.co/Qwen/Qwen3-TTS-12Hz-1.7B-VoiceDesign)\n- [Qwen/Qwen3-TTS-12Hz-1.7B-Base](https://huggingface.co/Qwen/Qwen3-TTS-12Hz-1.7B-Base)\n- [Qwen/Qwen3-TTS-12Hz-0.6B-Base](https://huggingface.co/Qwen/Qwen3-TTS-12Hz-0.6B-Base)\n- [Qwen/Qwen3-TTS-Tokenizer-12Hz](https://huggingface.co/Qwen/Qwen3-TTS-Tokenizer-12Hz)\n\n\n## FlashAttention (Optional)\n\nIf you have a compatible NVIDIA GPU, FlashAttention 2 can improve speed.\n\n```bash\npip install flash-attn --no-build-isolation\n```"],"color":"#432","bgcolor":"#653"},{"id":50,"type":"SaveAudioMP3","pos":[737.700114583396,1258.2181644307025],"size":[340,136],"flags":{},"order":6,"mode":4,"inputs":[{"localized_name":"音频","name":"audio","type":"AUDIO","link":36},{"localized_name":"文件名前缀","name":"filename_prefix","type":"STRING","widget":{"name":"filename_prefix"},"link":null},{"localized_name":"质量","name":"quality","type":"COMBO","widget":{"name":"quality"},"link":null},{"localized_name":"audioUI","name":"audioUI","type":"AUDIO_UI","widget":{"name":"audioUI"},"link":null}],"outputs":[],"properties":{"cnr_id":"comfy-core","ver":"0.10.0","Node name for S&R":"SaveAudioMP3","ue_properties":{"widget_ue_connectable":{},"input_ue_unconnectable":{}}},"widgets_values":["audio/ComfyUI","128k"]}],"links":[[35,47,0,46,0,"AUDIO"],[36,41,0,50,0,"AUDIO"],[37,46,0,51,0,"AUDIO"],[38,45,0,49,0,"AUDIO"]],"groups":[{"id":1,"title":"Voice Clone","bounding":[-7.841998505896003,694.2369715781095,360,711.8041660510136],"color":"#3f789e","font_size":24,"flags":{}},{"id":2,"title":"Custom  Voice","bounding":[358.7138198400269,873.8393213149233,359.99999999999994,532.2018163141998],"color":"#3f789e","font_size":24,"flags":{}},{"id":3,"title":"Voice Design","bounding":[727.700114583396,872.0163481165029,360,532.2018163141998],"color":"#3f789e","font_size":24,"flags":{}}],"config":{},"extra":{"workflowRendererVersion":"LG","ue_links":[],"links_added_by_ue":[],"ds":{"scale":0.9852511601783196,"offset":[721.5954161426612,-520.9225089114927]},"frontendVersion":"1.37.11","VHS_latentpreview":false,"VHS_latentpreviewrate":0,"VHS_MetadataImage":true,"VHS_KeepIntermediate":true},"version":0.4}

Additional Context

(Please add any additional context or steps to reproduce the error here)

Originally created by @DNPMBHC on GitHub (Jan 27, 2026). Original GitHub issue: https://github.com/1038lab/ComfyUI-QwenTTS/issues/10 # ComfyUI Error Report ## Error Details - **Node ID:** 45 - **Node Type:** AILab_Qwen3TTSCustomVoice - **Exception Type:** AttributeError - **Exception Message:** 'Qwen3TTSTalkerConfig' object has no attribute 'pad_token_id' ## Stack Trace ``` File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\execution.py", line 518, in execute output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\execution.py", line 329, in get_output_data return_values = await _async_map_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\execution.py", line 303, in _async_map_node_over_list await process_inputs(input_dict, i) File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\execution.py", line 291, in process_inputs result = f(**inputs) ^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\AILab_QwenTTS.py", line 554, in generate return _custom_voice_generate( ^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\AILab_QwenTTS.py", line 419, in _custom_voice_generate model = _load_model("CustomVoice", model_size, device, precision) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\AILab_QwenTTS.py", line 317, in _load_model model = Qwen3TTSModel.from_pretrained(source, device_map=device, dtype=dtype) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\inference\qwen3_tts_model.py", line 112, in from_pretrained model = AutoModel.from_pretrained(pretrained_model_name_or_path, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\models\auto\auto_factory.py", line 372, in from_pretrained return model_class.from_pretrained( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1872, in from_pretrained model = super().from_pretrained( ^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 4072, in from_pretrained model = cls(config, *model_args, **model_kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1821, in __init__ self.talker = Qwen3TTSTalkerForConditionalGeneration(self.config.talker_config) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1574, in __init__ self.model = Qwen3TTSTalkerModel(config) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1434, in __init__ self.padding_idx = config.pad_token_id ^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\configuration_utils.py", line 164, in __getattribute__ return super().__getattribute__(key) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ``` ## System Information - **ComfyUI Version:** 0.11.0 - **Arguments:** ComfyUI\main.py --windows-standalone-build - **OS:** win32 - **Python Version:** 3.12.10 (tags/v3.12.10:0cc8128, Apr 8 2025, 12:21:36) [MSC v.1943 64 bit (AMD64)] - **Embedded Python:** true - **PyTorch Version:** 2.11.0.dev20260115+cu130 ## Devices - **Name:** cuda:0 NVIDIA GeForce RTX 5060 Ti : cudaMallocAsync - **Type:** cuda - **VRAM Total:** 17102864384 - **VRAM Free:** 15896412160 - **Torch VRAM Total:** 0 - **Torch VRAM Free:** 0 ## Logs ``` 2026-01-27T18:51:29.948942 - [START] Security scan2026-01-27T18:51:29.948942 - 2026-01-27T18:51:32.286113 - [DONE] Security scan2026-01-27T18:51:32.286113 - 2026-01-27T18:51:32.592772 - ## ComfyUI-Manager: installing dependencies done.2026-01-27T18:51:32.592772 - 2026-01-27T18:51:32.592772 - ** ComfyUI startup time:2026-01-27T18:51:32.592772 - 2026-01-27T18:51:32.592772 - 2026-01-27 18:51:32.5922026-01-27T18:51:32.592772 - 2026-01-27T18:51:32.592772 - ** Platform:2026-01-27T18:51:32.593747 - 2026-01-27T18:51:32.593747 - Windows2026-01-27T18:51:32.593747 - 2026-01-27T18:51:32.593747 - ** Python version:2026-01-27T18:51:32.593747 - 2026-01-27T18:51:32.593747 - 3.12.10 (tags/v3.12.10:0cc8128, Apr 8 2025, 12:21:36) [MSC v.1943 64 bit (AMD64)]2026-01-27T18:51:32.593747 - 2026-01-27T18:51:32.593747 - ** Python executable:2026-01-27T18:51:32.593747 - 2026-01-27T18:51:32.593747 - E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\python.exe2026-01-27T18:51:32.593747 - 2026-01-27T18:51:32.594747 - ** ComfyUI Path:2026-01-27T18:51:32.594747 - 2026-01-27T18:51:32.594747 - E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI2026-01-27T18:51:32.594747 - 2026-01-27T18:51:32.594747 - ** ComfyUI Base Folder Path:2026-01-27T18:51:32.594747 - 2026-01-27T18:51:32.594747 - E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI2026-01-27T18:51:32.594747 - 2026-01-27T18:51:32.594747 - ** User directory:2026-01-27T18:51:32.594747 - 2026-01-27T18:51:32.594747 - E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\user2026-01-27T18:51:32.602906 - 2026-01-27T18:51:32.604000 - ** ComfyUI-Manager config path:2026-01-27T18:51:32.604000 - 2026-01-27T18:51:32.604000 - E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\user\__manager\config.ini2026-01-27T18:51:32.604000 - 2026-01-27T18:51:32.604000 - ** Log path:2026-01-27T18:51:32.604000 - 2026-01-27T18:51:32.604000 - E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\user\comfyui.log2026-01-27T18:51:32.604000 - 2026-01-27T18:51:35.026670 - Prestartup times for custom nodes: 2026-01-27T18:51:35.026670 - 5.6 seconds: E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-manager 2026-01-27T18:51:35.026670 - 2026-01-27T18:51:35.760805 - E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\cuda\__init__.py:61: FutureWarning: The pynvml package is deprecated. Please install nvidia-ml-py instead. If you did not install pynvml directly, please report this to the maintainers of the package that installed pynvml for you. import pynvml # type: ignore[import] 2026-01-27T18:51:38.886250 - Checkpoint files will always be loaded safely. 2026-01-27T18:51:39.029780 - Total VRAM 16311 MB, total RAM 65434 MB 2026-01-27T18:51:39.029780 - pytorch version: 2.11.0.dev20260115+cu130 2026-01-27T18:51:39.030778 - Set vram state to: NORMAL_VRAM 2026-01-27T18:51:39.031780 - Device: cuda:0 NVIDIA GeForce RTX 5060 Ti : cudaMallocAsync 2026-01-27T18:51:39.052881 - Using async weight offloading with 2 streams 2026-01-27T18:51:39.054386 - Enabled pinned memory 29445.0 2026-01-27T18:51:41.527823 - Found comfy_kitchen backend cuda: {'available': True, 'disabled': False, 'unavailable_reason': None, 'capabilities': ['apply_rope', 'apply_rope1', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8', 'scaled_mm_nvfp4']} 2026-01-27T18:51:41.528861 - Found comfy_kitchen backend eager: {'available': True, 'disabled': False, 'unavailable_reason': None, 'capabilities': ['apply_rope', 'apply_rope1', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8', 'scaled_mm_nvfp4']} 2026-01-27T18:51:41.528861 - Found comfy_kitchen backend triton: {'available': True, 'disabled': True, 'unavailable_reason': None, 'capabilities': ['apply_rope', 'apply_rope1', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8']} 2026-01-27T18:51:42.156347 - Using pytorch attention 2026-01-27T18:51:48.382001 - Python version: 3.12.10 (tags/v3.12.10:0cc8128, Apr 8 2025, 12:21:36) [MSC v.1943 64 bit (AMD64)] 2026-01-27T18:51:48.382001 - ComfyUI version: 0.11.0 2026-01-27T18:51:48.508878 - ComfyUI frontend version: 1.37.11 2026-01-27T18:51:48.512888 - [Prompt Server] web root: E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\comfyui_frontend_package\static 2026-01-27T18:51:50.585924 - [Crystools INFO] Crystools version: 1.27.4 2026-01-27T18:51:50.640074 - [Crystools INFO] Platform release: 11 2026-01-27T18:51:50.640074 - [Crystools INFO] JETSON: Not detected. 2026-01-27T18:51:50.642072 - [Crystools INFO] CPU: Intel(R) Core(TM) i3-10105 CPU @ 3.70GHz - Arch: AMD64 - OS: Windows 11 2026-01-27T18:51:50.654726 - [Crystools INFO] pynvml (NVIDIA) initialized. 2026-01-27T18:51:50.655728 - [Crystools INFO] GPU/s: 2026-01-27T18:51:50.673379 - [Crystools INFO] 0) NVIDIA GeForce RTX 5060 Ti 2026-01-27T18:51:50.673379 - [Crystools INFO] NVIDIA Driver: 591.74 2026-01-27T18:51:50.696692 - ### Loading: ComfyUI-Manager (V3.39.2) 2026-01-27T18:51:50.697702 - [ComfyUI-Manager] network_mode: public 2026-01-27T18:51:50.697702 - [ComfyUI-Manager] ComfyUI per-queue preview override detected (PR #11261). Manager's preview method feature is disabled. Use ComfyUI's --preview-method CLI option or 'Settings > Execution > Live preview method'. 2026-01-27T18:51:51.130856 - ### ComfyUI Version: v0.11.0 | Released on '2026-01-26' 2026-01-27T18:51:51.836698 - Warning: Unable to access E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\models\diffusion_models\wan2.2_i2v_14B_fp8_scaled. Skipping this path. 2026-01-27T18:51:51.837708 - Warning: Unable to access E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\models\diffusion_models\Wan2_2-T2V-A14B fp8. Skipping this path. 2026-01-27T18:51:51.875093 - Import times for custom nodes: 2026-01-27T18:51:51.876093 - 0.0 seconds: E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\websocket_image_save.py 2026-01-27T18:51:51.876093 - 0.0 seconds: E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-ReservedVRAM-main 2026-01-27T18:51:51.876093 - 0.0 seconds: E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-SageAttention3 2026-01-27T18:51:51.876093 - 0.0 seconds: E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS 2026-01-27T18:51:51.876093 - 0.0 seconds: E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-various-main 2026-01-27T18:51:51.877093 - 0.1 seconds: E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Crystools-main 2026-01-27T18:51:51.877093 - 0.5 seconds: E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-manager 2026-01-27T18:51:51.877093 - 0.7 seconds: E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_Swwan 2026-01-27T18:51:51.877093 - 2026-01-27T18:51:51.886142 - Context impl SQLiteImpl. 2026-01-27T18:51:51.886142 - Will assume non-transactional DDL. 2026-01-27T18:51:52.079533 - Assets scan(roots=['models']) completed in 0.185s (created=0, skipped_existing=111, total_seen=122) 2026-01-27T18:51:52.127119 - [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/alter-list.json 2026-01-27T18:51:52.300139 - Starting server 2026-01-27T18:51:52.301107 - To see the GUI go to: http://127.0.0.1:8188 2026-01-27T18:51:52.337612 - [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/model-list.json 2026-01-27T18:51:52.526812 - [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/github-stats.json 2026-01-27T18:51:52.944291 - [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/extension-node-map.json 2026-01-27T18:51:53.527647 - [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/custom-node-list.json 2026-01-27T18:51:53.770576 - [DEPRECATION WARNING] Detected import of deprecated legacy API: /scripts/ui/components/buttonGroup.js. This is likely caused by a custom node extension using outdated APIs. Please update your extensions or contact the extension author for an updated version. 2026-01-27T18:51:53.772592 - [DEPRECATION WARNING] Detected import of deprecated legacy API: /scripts/ui.js. This is likely caused by a custom node extension using outdated APIs. Please update your extensions or contact the extension author for an updated version. 2026-01-27T18:51:53.773592 - [DEPRECATION WARNING] Detected import of deprecated legacy API: /extensions/core/groupNode.js. This is likely caused by a custom node extension using outdated APIs. Please update your extensions or contact the extension author for an updated version. 2026-01-27T18:51:54.861922 - [DEPRECATION WARNING] Detected import of deprecated legacy API: /scripts/ui/components/button.js. This is likely caused by a custom node extension using outdated APIs. Please update your extensions or contact the extension author for an updated version. 2026-01-27T18:51:58.164217 - got prompt 2026-01-27T18:51:59.586616 - speaker_encoder_config is None. Initializing talker model with default values 2026-01-27T18:51:59.587654 - E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_rope_utils.py:927: FutureWarning: `rope_config_validation` is deprecated and has been removed. Its functionality has been moved to RotaryEmbeddingConfigMixin.validate_rope method. PreTrainedConfig inherits this class, so please call self.validate_rope() instead. Also, make sure to use the new rope_parameters syntax. You can call self.standardize_rope_params() in the meantime. warnings.warn( 2026-01-27T18:51:59.589161 - talker_config is None. Initializing talker model with default values 2026-01-27T18:51:59.589161 - speaker_encoder_config is None. Initializing talker model with default values 2026-01-27T18:51:59.590171 - code_predictor_config is None. Initializing code_predictor model with default values 2026-01-27T18:51:59.591169 - code_predictor_config is None. Initializing code_predictor model with default values 2026-01-27T18:51:59.595166 - [Qwen3-TTS] flash_attention_2 unavailable, fallback to default attention2026-01-27T18:51:59.595166 - 2026-01-27T18:51:59.598682 - speaker_encoder_config is None. Initializing talker model with default values 2026-01-27T18:51:59.599680 - talker_config is None. Initializing talker model with default values 2026-01-27T18:51:59.599680 - speaker_encoder_config is None. Initializing talker model with default values 2026-01-27T18:51:59.599680 - code_predictor_config is None. Initializing code_predictor model with default values 2026-01-27T18:51:59.600679 - code_predictor_config is None. Initializing code_predictor model with default values 2026-01-27T18:51:59.607833 - !!! Exception during processing !!! 'Qwen3TTSTalkerConfig' object has no attribute 'pad_token_id' 2026-01-27T18:51:59.621879 - Traceback (most recent call last): File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\AILab_QwenTTS.py", line 311, in _load_model model = Qwen3TTSModel.from_pretrained(source, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\inference\qwen3_tts_model.py", line 112, in from_pretrained model = AutoModel.from_pretrained(pretrained_model_name_or_path, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\models\auto\auto_factory.py", line 372, in from_pretrained return model_class.from_pretrained( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1872, in from_pretrained model = super().from_pretrained( ^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 4072, in from_pretrained model = cls(config, *model_args, **model_kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1818, in __init__ super().__init__(config) File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 1295, in __init__ self.config._attn_implementation_internal = self._check_and_adjust_attn_implementation( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 1873, in _check_and_adjust_attn_implementation applicable_attn_implementation = self.get_correct_attn_implementation( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 1914, in get_correct_attn_implementation self._flash_attn_2_can_dispatch(is_init_check) File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 1594, in _flash_attn_2_can_dispatch raise ImportError(f"{preface} the package flash_attn seems to be not installed. {install_message}") ImportError: FlashAttention2 has been toggled on, but it cannot be used due to the following error: the package flash_attn seems to be not installed. Please refer to the documentation of https://huggingface.co/docs/transformers/perf_infer_gpu_one#flashattention-2 to install Flash Attention 2. During handling of the above exception, another exception occurred: Traceback (most recent call last): File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\execution.py", line 518, in execute output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\execution.py", line 329, in get_output_data return_values = await _async_map_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\execution.py", line 303, in _async_map_node_over_list await process_inputs(input_dict, i) File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\execution.py", line 291, in process_inputs result = f(**inputs) ^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\AILab_QwenTTS.py", line 644, in generate return _voice_design_generate( ^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\AILab_QwenTTS.py", line 460, in _voice_design_generate model = _load_model("VoiceDesign", model_size, device, precision) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\AILab_QwenTTS.py", line 317, in _load_model model = Qwen3TTSModel.from_pretrained(source, device_map=device, dtype=dtype) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\inference\qwen3_tts_model.py", line 112, in from_pretrained model = AutoModel.from_pretrained(pretrained_model_name_or_path, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\models\auto\auto_factory.py", line 372, in from_pretrained return model_class.from_pretrained( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1872, in from_pretrained model = super().from_pretrained( ^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 4072, in from_pretrained model = cls(config, *model_args, **model_kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1821, in __init__ self.talker = Qwen3TTSTalkerForConditionalGeneration(self.config.talker_config) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1574, in __init__ self.model = Qwen3TTSTalkerModel(config) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1434, in __init__ self.padding_idx = config.pad_token_id ^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\configuration_utils.py", line 164, in __getattribute__ return super().__getattribute__(key) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ AttributeError: 'Qwen3TTSTalkerConfig' object has no attribute 'pad_token_id' 2026-01-27T18:51:59.627399 - Prompt executed in 1.46 seconds 2026-01-27T18:52:02.534882 - FETCH ComfyRegistry Data: 5/1222026-01-27T18:52:02.534882 - 2026-01-27T18:52:11.762719 - FETCH ComfyRegistry Data: 10/1222026-01-27T18:52:11.763386 - 2026-01-27T18:52:20.401999 - FETCH ComfyRegistry Data: 15/1222026-01-27T18:52:20.401999 - 2026-01-27T18:52:28.744270 - FETCH ComfyRegistry Data: 20/1222026-01-27T18:52:28.745269 - 2026-01-27T18:52:36.897609 - FETCH ComfyRegistry Data: 25/1222026-01-27T18:52:36.897609 - 2026-01-27T18:52:46.058172 - FETCH ComfyRegistry Data: 30/1222026-01-27T18:52:46.058676 - 2026-01-27T18:52:55.796253 - FETCH ComfyRegistry Data: 35/1222026-01-27T18:52:55.796253 - 2026-01-27T18:53:03.980159 - FETCH ComfyRegistry Data: 40/1222026-01-27T18:53:03.982185 - 2026-01-27T18:53:12.278949 - FETCH ComfyRegistry Data: 45/1222026-01-27T18:53:12.278949 - 2026-01-27T18:53:21.079325 - FETCH ComfyRegistry Data: 50/1222026-01-27T18:53:21.079998 - 2026-01-27T18:53:30.454535 - FETCH ComfyRegistry Data: 55/1222026-01-27T18:53:30.454535 - 2026-01-27T18:53:38.713102 - FETCH ComfyRegistry Data: 60/1222026-01-27T18:53:38.714131 - 2026-01-27T18:53:48.433411 - FETCH ComfyRegistry Data: 65/1222026-01-27T18:53:48.433411 - 2026-01-27T18:53:58.045134 - FETCH ComfyRegistry Data: 70/1222026-01-27T18:53:58.045134 - 2026-01-27T18:54:08.422011 - FETCH ComfyRegistry Data: 75/1222026-01-27T18:54:08.422524 - 2026-01-27T18:54:17.137583 - FETCH ComfyRegistry Data: 80/1222026-01-27T18:54:17.137583 - 2026-01-27T18:54:25.472192 - FETCH ComfyRegistry Data: 85/1222026-01-27T18:54:25.472192 - 2026-01-27T18:54:34.968890 - FETCH ComfyRegistry Data: 90/1222026-01-27T18:54:34.968890 - 2026-01-27T18:54:43.334997 - FETCH ComfyRegistry Data: 95/1222026-01-27T18:54:43.334997 - 2026-01-27T18:54:56.398747 - FETCH ComfyRegistry Data: 100/1222026-01-27T18:54:56.398747 - 2026-01-27T18:55:05.589262 - FETCH ComfyRegistry Data: 105/1222026-01-27T18:55:05.589262 - 2026-01-27T18:55:14.273031 - FETCH ComfyRegistry Data: 110/1222026-01-27T18:55:14.273031 - 2026-01-27T18:55:22.980476 - FETCH ComfyRegistry Data: 115/1222026-01-27T18:55:22.980476 - 2026-01-27T18:55:31.563040 - FETCH ComfyRegistry Data: 120/1222026-01-27T18:55:31.563040 - 2026-01-27T18:55:35.089320 - FETCH ComfyRegistry Data [DONE]2026-01-27T18:55:35.091349 - 2026-01-27T18:55:35.363256 - [ComfyUI-Manager] default cache updated: https://api.comfy.org/nodes 2026-01-27T18:55:35.400501 - FETCH DATA from: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/custom-node-list.json2026-01-27T18:55:37.166808 - [DONE]2026-01-27T18:55:37.167834 - 2026-01-27T18:55:37.405881 - [ComfyUI-Manager] All startup tasks have been completed. 2026-01-27T18:57:24.247018 - got prompt 2026-01-27T18:57:24.268084 - speaker_encoder_config is None. Initializing talker model with default values 2026-01-27T18:57:24.269595 - talker_config is None. Initializing talker model with default values 2026-01-27T18:57:24.270607 - speaker_encoder_config is None. Initializing talker model with default values 2026-01-27T18:57:24.272632 - code_predictor_config is None. Initializing code_predictor model with default values 2026-01-27T18:57:24.274626 - code_predictor_config is None. Initializing code_predictor model with default values 2026-01-27T18:57:24.286181 - [Qwen3-TTS] flash_attention_2 unavailable, fallback to default attention2026-01-27T18:57:24.287194 - 2026-01-27T18:57:24.293098 - speaker_encoder_config is None. Initializing talker model with default values 2026-01-27T18:57:24.294875 - talker_config is None. Initializing talker model with default values 2026-01-27T18:57:24.294875 - speaker_encoder_config is None. Initializing talker model with default values 2026-01-27T18:57:24.295421 - code_predictor_config is None. Initializing code_predictor model with default values 2026-01-27T18:57:24.295963 - code_predictor_config is None. Initializing code_predictor model with default values 2026-01-27T18:57:24.307589 - !!! Exception during processing !!! 'Qwen3TTSTalkerConfig' object has no attribute 'pad_token_id' 2026-01-27T18:57:24.314627 - Traceback (most recent call last): File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\AILab_QwenTTS.py", line 311, in _load_model model = Qwen3TTSModel.from_pretrained(source, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\inference\qwen3_tts_model.py", line 112, in from_pretrained model = AutoModel.from_pretrained(pretrained_model_name_or_path, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\models\auto\auto_factory.py", line 372, in from_pretrained return model_class.from_pretrained( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1872, in from_pretrained model = super().from_pretrained( ^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 4072, in from_pretrained model = cls(config, *model_args, **model_kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1818, in __init__ super().__init__(config) File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 1295, in __init__ self.config._attn_implementation_internal = self._check_and_adjust_attn_implementation( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 1873, in _check_and_adjust_attn_implementation applicable_attn_implementation = self.get_correct_attn_implementation( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 1914, in get_correct_attn_implementation self._flash_attn_2_can_dispatch(is_init_check) File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 1594, in _flash_attn_2_can_dispatch raise ImportError(f"{preface} the package flash_attn seems to be not installed. {install_message}") ImportError: FlashAttention2 has been toggled on, but it cannot be used due to the following error: the package flash_attn seems to be not installed. Please refer to the documentation of https://huggingface.co/docs/transformers/perf_infer_gpu_one#flashattention-2 to install Flash Attention 2. During handling of the above exception, another exception occurred: Traceback (most recent call last): File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\execution.py", line 518, in execute output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\execution.py", line 329, in get_output_data return_values = await _async_map_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\execution.py", line 303, in _async_map_node_over_list await process_inputs(input_dict, i) File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\execution.py", line 291, in process_inputs result = f(**inputs) ^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\AILab_QwenTTS.py", line 554, in generate return _custom_voice_generate( ^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\AILab_QwenTTS.py", line 419, in _custom_voice_generate model = _load_model("CustomVoice", model_size, device, precision) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\AILab_QwenTTS.py", line 317, in _load_model model = Qwen3TTSModel.from_pretrained(source, device_map=device, dtype=dtype) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\inference\qwen3_tts_model.py", line 112, in from_pretrained model = AutoModel.from_pretrained(pretrained_model_name_or_path, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\models\auto\auto_factory.py", line 372, in from_pretrained return model_class.from_pretrained( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1872, in from_pretrained model = super().from_pretrained( ^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 4072, in from_pretrained model = cls(config, *model_args, **model_kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1821, in __init__ self.talker = Qwen3TTSTalkerForConditionalGeneration(self.config.talker_config) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1574, in __init__ self.model = Qwen3TTSTalkerModel(config) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1434, in __init__ self.padding_idx = config.pad_token_id ^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\configuration_utils.py", line 164, in __getattribute__ return super().__getattribute__(key) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ AttributeError: 'Qwen3TTSTalkerConfig' object has no attribute 'pad_token_id' 2026-01-27T18:57:24.320144 - Prompt executed in 0.07 seconds 2026-01-27T18:58:29.028882 - got prompt 2026-01-27T18:58:29.046458 - speaker_encoder_config is None. Initializing talker model with default values 2026-01-27T18:58:29.048771 - talker_config is None. Initializing talker model with default values 2026-01-27T18:58:29.048771 - speaker_encoder_config is None. Initializing talker model with default values 2026-01-27T18:58:29.048771 - code_predictor_config is None. Initializing code_predictor model with default values 2026-01-27T18:58:29.049796 - code_predictor_config is None. Initializing code_predictor model with default values 2026-01-27T18:58:29.054863 - [Qwen3-TTS] flash_attention_2 unavailable, fallback to default attention2026-01-27T18:58:29.054863 - 2026-01-27T18:58:29.064403 - speaker_encoder_config is None. Initializing talker model with default values 2026-01-27T18:58:29.068926 - talker_config is None. Initializing talker model with default values 2026-01-27T18:58:29.069937 - speaker_encoder_config is None. Initializing talker model with default values 2026-01-27T18:58:29.071946 - code_predictor_config is None. Initializing code_predictor model with default values 2026-01-27T18:58:29.072948 - code_predictor_config is None. Initializing code_predictor model with default values 2026-01-27T18:58:29.085010 - !!! Exception during processing !!! 'Qwen3TTSTalkerConfig' object has no attribute 'pad_token_id' 2026-01-27T18:58:29.094118 - Traceback (most recent call last): File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\AILab_QwenTTS.py", line 311, in _load_model model = Qwen3TTSModel.from_pretrained(source, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\inference\qwen3_tts_model.py", line 112, in from_pretrained model = AutoModel.from_pretrained(pretrained_model_name_or_path, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\models\auto\auto_factory.py", line 372, in from_pretrained return model_class.from_pretrained( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1872, in from_pretrained model = super().from_pretrained( ^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 4072, in from_pretrained model = cls(config, *model_args, **model_kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1818, in __init__ super().__init__(config) File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 1295, in __init__ self.config._attn_implementation_internal = self._check_and_adjust_attn_implementation( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 1873, in _check_and_adjust_attn_implementation applicable_attn_implementation = self.get_correct_attn_implementation( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 1914, in get_correct_attn_implementation self._flash_attn_2_can_dispatch(is_init_check) File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 1594, in _flash_attn_2_can_dispatch raise ImportError(f"{preface} the package flash_attn seems to be not installed. {install_message}") ImportError: FlashAttention2 has been toggled on, but it cannot be used due to the following error: the package flash_attn seems to be not installed. Please refer to the documentation of https://huggingface.co/docs/transformers/perf_infer_gpu_one#flashattention-2 to install Flash Attention 2. During handling of the above exception, another exception occurred: Traceback (most recent call last): File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\execution.py", line 518, in execute output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\execution.py", line 329, in get_output_data return_values = await _async_map_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\execution.py", line 303, in _async_map_node_over_list await process_inputs(input_dict, i) File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\execution.py", line 291, in process_inputs result = f(**inputs) ^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\AILab_QwenTTS.py", line 554, in generate return _custom_voice_generate( ^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\AILab_QwenTTS.py", line 419, in _custom_voice_generate model = _load_model("CustomVoice", model_size, device, precision) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\AILab_QwenTTS.py", line 317, in _load_model model = Qwen3TTSModel.from_pretrained(source, device_map=device, dtype=dtype) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\inference\qwen3_tts_model.py", line 112, in from_pretrained model = AutoModel.from_pretrained(pretrained_model_name_or_path, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\models\auto\auto_factory.py", line 372, in from_pretrained return model_class.from_pretrained( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1872, in from_pretrained model = super().from_pretrained( ^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 4072, in from_pretrained model = cls(config, *model_args, **model_kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1821, in __init__ self.talker = Qwen3TTSTalkerForConditionalGeneration(self.config.talker_config) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1574, in __init__ self.model = Qwen3TTSTalkerModel(config) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QwenTTS\qwen_tts\core\models\modeling_qwen3_tts.py", line 1434, in __init__ self.padding_idx = config.pad_token_id ^^^^^^^^^^^^^^^^^^^ File "E:\Comfyui_CUDA_13\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\configuration_utils.py", line 164, in __getattribute__ return super().__getattribute__(key) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ AttributeError: 'Qwen3TTSTalkerConfig' object has no attribute 'pad_token_id' 2026-01-27T18:58:29.098395 - Prompt executed in 0.07 seconds ``` ## Attached Workflow Please make sure that workflow does not contain any sensitive information such as API keys or passwords. ``` {"id":"6b2e5d09-d1ea-4fc9-a8bd-8daca49d65ff","revision":0,"last_node_id":52,"last_link_id":38,"nodes":[{"id":46,"type":"AILab_Qwen3TTSVoiceClone","pos":[2.1580014941039964,947.4393213149233],"size":[340,262],"flags":{},"order":4,"mode":4,"inputs":[{"localized_name":"reference_audio","name":"reference_audio","type":"AUDIO","link":35},{"localized_name":"target_text","name":"target_text","type":"STRING","widget":{"name":"target_text"},"link":null},{"localized_name":"model_size","name":"model_size","type":"COMBO","widget":{"name":"model_size"},"link":null},{"localized_name":"language","name":"language","type":"COMBO","widget":{"name":"language"},"link":null},{"localized_name":"reference_text","name":"reference_text","shape":7,"type":"STRING","widget":{"name":"reference_text"},"link":null},{"localized_name":"x_vector_only","name":"x_vector_only","shape":7,"type":"BOOLEAN","widget":{"name":"x_vector_only"},"link":null},{"localized_name":"seed","name":"seed","shape":7,"type":"INT","widget":{"name":"seed"},"link":null}],"outputs":[{"localized_name":"音频","name":"AUDIO","type":"AUDIO","links":[37]}],"properties":{"cnr_id":"comfyui-qwentts","ver":"1.0.1","Node name for S&R":"AILab_Qwen3TTSVoiceClone","ue_properties":{"widget_ue_connectable":{},"input_ue_unconnectable":{},"version":"7.5.2"}},"widgets_values":["Hello, this is a cloned voice.","0.6B","Auto","",false,875958189094423,"randomize"],"color":"#4D2600","bgcolor":"#3e5249"},{"id":47,"type":"LoadAudio","pos":[2.1580014941039964,767.8369715781096],"size":[340,136],"flags":{},"order":0,"mode":4,"inputs":[{"localized_name":"音频","name":"audio","type":"COMBO","widget":{"name":"audio"},"link":null},{"localized_name":"音频UI","name":"audioUI","type":"AUDIO_UI","widget":{"name":"audioUI"},"link":null},{"localized_name":"选择文件上传","name":"upload","type":"AUDIOUPLOAD","widget":{"name":"upload"},"link":null}],"outputs":[{"localized_name":"音频","name":"AUDIO","type":"AUDIO","links":[35]}],"properties":{"cnr_id":"comfy-core","ver":"0.10.0","Node name for S&R":"LoadAudio","ue_properties":{"widget_ue_connectable":{},"input_ue_unconnectable":{}}},"widgets_values":["107289363.mp4",null,null]},{"id":51,"type":"SaveAudioMP3","pos":[2.1580014941039964,1260.041137629123],"size":[340,136],"flags":{},"order":7,"mode":4,"inputs":[{"localized_name":"音频","name":"audio","type":"AUDIO","link":37},{"localized_name":"文件名前缀","name":"filename_prefix","type":"STRING","widget":{"name":"filename_prefix"},"link":null},{"localized_name":"质量","name":"quality","type":"COMBO","widget":{"name":"quality"},"link":null},{"localized_name":"audioUI","name":"audioUI","type":"AUDIO_UI","widget":{"name":"audioUI"},"link":null}],"outputs":[],"properties":{"cnr_id":"comfy-core","ver":"0.10.0","Node name for S&R":"SaveAudioMP3","ue_properties":{"widget_ue_connectable":{},"input_ue_unconnectable":{}}},"widgets_values":["audio/ComfyUI","320k"]},{"id":49,"type":"SaveAudioMP3","pos":[368.7138198400269,1260.041137629123],"size":[340,136],"flags":{},"order":5,"mode":0,"inputs":[{"localized_name":"音频","name":"audio","type":"AUDIO","link":38},{"localized_name":"文件名前缀","name":"filename_prefix","type":"STRING","widget":{"name":"filename_prefix"},"link":null},{"localized_name":"质量","name":"quality","type":"COMBO","widget":{"name":"quality"},"link":null},{"localized_name":"audioUI","name":"audioUI","type":"AUDIO_UI","widget":{"name":"audioUI"},"link":null}],"outputs":[],"properties":{"cnr_id":"comfy-core","ver":"0.10.0","Node name for S&R":"SaveAudioMP3","ue_properties":{"widget_ue_connectable":{},"input_ue_unconnectable":{}}},"widgets_values":["audio/ComfyUI","V0"]},{"id":45,"type":"AILab_Qwen3TTSCustomVoice","pos":[368.7138198400269,947.4393213149233],"size":[340,262],"flags":{},"order":1,"mode":0,"inputs":[{"localized_name":"text","name":"text","type":"STRING","widget":{"name":"text"},"link":null},{"localized_name":"speaker","name":"speaker","type":"COMBO","widget":{"name":"speaker"},"link":null},{"localized_name":"model_size","name":"model_size","type":"COMBO","widget":{"name":"model_size"},"link":null},{"localized_name":"language","name":"language","type":"COMBO","widget":{"name":"language"},"link":null},{"localized_name":"instruct","name":"instruct","shape":7,"type":"STRING","widget":{"name":"instruct"},"link":null},{"localized_name":"seed","name":"seed","shape":7,"type":"INT","widget":{"name":"seed"},"link":null}],"outputs":[{"localized_name":"音频","name":"AUDIO","type":"AUDIO","links":[38]}],"properties":{"cnr_id":"comfyui-qwentts","ver":"1.0.1","Node name for S&R":"AILab_Qwen3TTSCustomVoice","ue_properties":{"widget_ue_connectable":{},"input_ue_unconnectable":{},"version":"7.5.2"}},"widgets_values":["Hello from Qwen3-TTS.","Ryan","1.7B","Auto","",342286630204760,"randomize"],"color":"#4D2600","bgcolor":"#3e5249"},{"id":41,"type":"AILab_Qwen3TTSVoiceDesign","pos":[737.700114583396,945.6163481165029],"size":[340,238],"flags":{},"order":2,"mode":4,"inputs":[{"localized_name":"text","name":"text","type":"STRING","widget":{"name":"text"},"link":null},{"localized_name":"instruct","name":"instruct","type":"STRING","widget":{"name":"instruct"},"link":null},{"localized_name":"model_size","name":"model_size","type":"COMBO","widget":{"name":"model_size"},"link":null},{"localized_name":"language","name":"language","type":"COMBO","widget":{"name":"language"},"link":null},{"localized_name":"seed","name":"seed","shape":7,"type":"INT","widget":{"name":"seed"},"link":null}],"outputs":[{"localized_name":"音频","name":"AUDIO","type":"AUDIO","links":[36]}],"properties":{"cnr_id":"comfyui-qwentts","ver":"1.0.1","Node name for S&R":"AILab_Qwen3TTSVoiceDesign","ue_properties":{"widget_ue_connectable":{},"input_ue_unconnectable":{},"version":"7.5.2"}},"widgets_values":["Hello from Qwen3-TTS VoiceDesign.","A warm, gentle female voice.","1.7B","Auto",242260717380428,"randomize"],"color":"#4D2600","bgcolor":"#3e5249"},{"id":52,"type":"MarkdownNote","pos":[-418.20320401527084,760.3485979843861],"size":[383.75722406036607,599.0649205241322],"flags":{},"order":3,"mode":0,"inputs":[],"outputs":[],"properties":{},"widgets_values":["# ComfyUI-QwenTTS\n\nA clean, efficient ComfyUI custom node pack for **Qwen3-TTS**. It provides **CustomVoice**, **VoiceDesign**, and **VoiceClone** workflows with strict ComfyUI compatibility and practical controls for quality, speed, and stability.\n\n## Usage\n\n### Custom Voice TTS\nAdd **Qwen3 TTS CustomVoice** and fill `text`, choose `speaker`, optionally add `instruct`.\n\n### Voice Design\nAdd **Qwen3 TTS VoiceDesign**, provide `text` and `instruct`.\n\n### Voice Clone\nAdd **Qwen3 TTS VoiceClone**, connect `reference_audio`, and provide `reference_text`.\n\n## Model Download\n\nModels can be auto-downloaded to:\n```\nComfyUI/models/TTS/Qwen3-TTS/<MODEL_NAME>/\n```\n\nSupported model IDs (Hugging Face):\n- [Qwen/Qwen3-TTS-12Hz-1.7B-CustomVoice](https://huggingface.co/Qwen/Qwen3-TTS-12Hz-1.7B-CustomVoice)\n- [Qwen/Qwen3-TTS-12Hz-0.6B-CustomVoice](https://huggingface.co/Qwen/Qwen3-TTS-12Hz-0.6B-CustomVoice)\n- [Qwen/Qwen3-TTS-12Hz-1.7B-VoiceDesign](https://huggingface.co/Qwen/Qwen3-TTS-12Hz-1.7B-VoiceDesign)\n- [Qwen/Qwen3-TTS-12Hz-1.7B-Base](https://huggingface.co/Qwen/Qwen3-TTS-12Hz-1.7B-Base)\n- [Qwen/Qwen3-TTS-12Hz-0.6B-Base](https://huggingface.co/Qwen/Qwen3-TTS-12Hz-0.6B-Base)\n- [Qwen/Qwen3-TTS-Tokenizer-12Hz](https://huggingface.co/Qwen/Qwen3-TTS-Tokenizer-12Hz)\n\n\n## FlashAttention (Optional)\n\nIf you have a compatible NVIDIA GPU, FlashAttention 2 can improve speed.\n\n```bash\npip install flash-attn --no-build-isolation\n```"],"color":"#432","bgcolor":"#653"},{"id":50,"type":"SaveAudioMP3","pos":[737.700114583396,1258.2181644307025],"size":[340,136],"flags":{},"order":6,"mode":4,"inputs":[{"localized_name":"音频","name":"audio","type":"AUDIO","link":36},{"localized_name":"文件名前缀","name":"filename_prefix","type":"STRING","widget":{"name":"filename_prefix"},"link":null},{"localized_name":"质量","name":"quality","type":"COMBO","widget":{"name":"quality"},"link":null},{"localized_name":"audioUI","name":"audioUI","type":"AUDIO_UI","widget":{"name":"audioUI"},"link":null}],"outputs":[],"properties":{"cnr_id":"comfy-core","ver":"0.10.0","Node name for S&R":"SaveAudioMP3","ue_properties":{"widget_ue_connectable":{},"input_ue_unconnectable":{}}},"widgets_values":["audio/ComfyUI","128k"]}],"links":[[35,47,0,46,0,"AUDIO"],[36,41,0,50,0,"AUDIO"],[37,46,0,51,0,"AUDIO"],[38,45,0,49,0,"AUDIO"]],"groups":[{"id":1,"title":"Voice Clone","bounding":[-7.841998505896003,694.2369715781095,360,711.8041660510136],"color":"#3f789e","font_size":24,"flags":{}},{"id":2,"title":"Custom Voice","bounding":[358.7138198400269,873.8393213149233,359.99999999999994,532.2018163141998],"color":"#3f789e","font_size":24,"flags":{}},{"id":3,"title":"Voice Design","bounding":[727.700114583396,872.0163481165029,360,532.2018163141998],"color":"#3f789e","font_size":24,"flags":{}}],"config":{},"extra":{"workflowRendererVersion":"LG","ue_links":[],"links_added_by_ue":[],"ds":{"scale":0.9852511601783196,"offset":[721.5954161426612,-520.9225089114927]},"frontendVersion":"1.37.11","VHS_latentpreview":false,"VHS_latentpreviewrate":0,"VHS_MetadataImage":true,"VHS_KeepIntermediate":true},"version":0.4} ``` ## Additional Context (Please add any additional context or steps to reproduce the error here)

bsakaguchi commented 2026-10-30 21:03:25 +00:00

Author

Owner

Copy link

@teslor commented on GitHub (Jan 27, 2026):

It seems it's an issue with transformers. For me this workaround helped.

<!-- gh-comment-id:3805744796 --> @teslor commented on GitHub (Jan 27, 2026): It seems it's an issue with transformers. For me [this](https://github.com/flybirdxx/ComfyUI-Qwen-TTS/issues/21#issuecomment-3795789903) workaround helped.

bsakaguchi commented 2026-10-30 21:03:25 +00:00

Author

Owner

Copy link

@1038lab commented on GitHub (Jan 30, 2026):

Likely caused by Transformers 5.x dev / incompatible Transformers build in the ComfyUI environment.

Draft reply (English):

Thanks for the detailed report + stacktrace — this error is almost always caused by an incompatible transformers version (often 5.0.0.dev0 / nightly) being installed in the ComfyUI Python environment.

Fix (recommended): pin Transformers to a stable 4.57.x

If you use ComfyUI portable / embedded python (Windows):

cd <your ComfyUI folder>
python_embeded\python.exe -m pip install -U "transformers==4.57.3" "tokenizers<0.20" --no-cache-dir
python_embeded\python.exe -m pip install -r ComfyUI\custom_nodes\ComfyUI-QwenTTS\requirements.txt --no-cache-dir

If you use system python:

pip install -U "transformers==4.57.3" "tokenizers<0.20" --no-cache-dir

Then restart ComfyUI.

If it still happens, please paste:

• pip show transformers tokenizers
• which model you selected (0.6B/1.7B)
• whether you installed via Manager or git clone

<!-- gh-comment-id:3821883170 --> @1038lab commented on GitHub (Jan 30, 2026): Likely caused by **Transformers 5.x dev / incompatible Transformers build** in the ComfyUI environment. **Draft reply (English):** Thanks for the detailed report + stacktrace — this error is almost always caused by an incompatible `transformers` version (often `5.0.0.dev0` / nightly) being installed in the ComfyUI Python environment. **Fix (recommended): pin Transformers to a stable 4.57.x** If you use ComfyUI portable / embedded python (Windows): ```bash cd <your ComfyUI folder> python_embeded\python.exe -m pip install -U "transformers==4.57.3" "tokenizers<0.20" --no-cache-dir python_embeded\python.exe -m pip install -r ComfyUI\custom_nodes\ComfyUI-QwenTTS\requirements.txt --no-cache-dir ``` If you use system python: ```bash pip install -U "transformers==4.57.3" "tokenizers<0.20" --no-cache-dir ``` Then **restart ComfyUI**. If it still happens, please paste: - `pip show transformers tokenizers` - which model you selected (0.6B/1.7B) - whether you installed via Manager or git clone

bsakaguchi commented 2026-10-30 21:03:26 +00:00

Author

Owner

Copy link

@codexiaoyu commented on GitHub (Feb 12, 2026):

The same issue also occurs with transformers==4.57.3 and tokenizers==0.22.1.
Does it mean that tokenizers < 0.20 is required for it to work properly?

<!-- gh-comment-id:3892001916 --> @codexiaoyu commented on GitHub (Feb 12, 2026): The same issue also occurs with transformers==4.57.3 and tokenizers==0.22.1. Does it mean that tokenizers < 0.20 is required for it to work properly?

bsakaguchi commented 2026-10-30 21:03:26 +00:00

Author

Owner

Copy link

@kynalvarus commented on GitHub (Sep 23, 2026):

If you use system python:

pip install -U "transformers==4.57.3" "tokenizers<0.20" --no-cache-dir
Then restart ComfyUI.

This is an unsolvable configuration.

No solution found when resolving dependencies:
  ╰─▶ Because transformers==4.57.6 depends on tokenizers>=0.22.0,<=0.23.0 and you require transformers==4.57.6, we can conclude that you require tokenizers>=0.22.0,<=0.23.0.

But pip install -U "transformers==4.57.6" "tokenizers<0.23" --no-cache-dir cleared it up.

<!-- gh-comment-id:5804485997 --> @kynalvarus commented on GitHub (Sep 23, 2026): > If you use system python: > > pip install -U "transformers==4.57.3" "tokenizers<0.20" --no-cache-dir > Then **restart ComfyUI**. > This is an unsolvable configuration. ``` No solution found when resolving dependencies: ╰─▶ Because transformers==4.57.6 depends on tokenizers>=0.22.0,<=0.23.0 and you require transformers==4.57.6, we can conclude that you require tokenizers>=0.22.0,<=0.23.0. ``` But `pip install -U "transformers==4.57.6" "tokenizers<0.23" --no-cache-dir` cleared it up.

bsakaguchi referenced this issue 2026-10-30 21:03:38 +00:00

[PR #21] Add compatibility for Transformers 5.x #22

Sign in to join this conversation.

No Branch/Tag specified

Branches Tags

main

No results found.

Labels

Clear labels   

pull-request

Mirrored from GitHub Pull Request

No labels

pull-request

Milestone

Clear milestone

No items

No milestone

Projects

Clear projects

No items

No project

Assignees

Clear assignees

No assignees

1 participant

Notifications

Subscribe

Due date

The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference

1038lab/ComfyUI-QwenTTS#9

No description provided.