[转载搬运] 剑侠AI-krea2文生图整合包v2.1，最低8G显卡，支持NSFW [1+50GB][百度盘]

静若

ZERO.零ꦿృ༊ 发表于 2026-7-22 20:06
启动超时，问了下ai说没有安装git这个插件是怎么回事

检查根目录，data/runtime.log日志报错文件，安装整合包必备前置：
Microsoft Visual C++ v14 Redistributable (x64)

1176200492

大佬，每次解压到最后一点的时候就弹出这个框是怎么回事呀

zhangyuanqi

我这个文生图脱衣有点失真，有没有提高真实度的提示词嘞

静若

1176200492 发表于 2026-7-29 20:10
大佬，每次解压到最后一点的时候就弹出这个框是怎么回事呀

1、优先排查有没有关闭杀毒软件，因为杀毒软件会拦截误报，导致解压失败
2、要么就是压缩包在下载过程损坏了或者没有被完整下载，解压也会失败

静若

yc87654321 发表于 2026-7-29 13:12
一直提示正在生成中，等了很久完全不出图啊？？

关闭后台无关的软件，留出显存，可提高生成速度，看看右下角终端有没有在动，还是卡住了，而不是盯着中间区域看ai生成！要学会排查问题，不懂的可以问AI，如豆包，deepseek等

1012827767771

大佬提示词有一点难有没有提供一点色情的提示词模板

1176200492

流弊大佬，换了个band解压软件就成功解压了，并且成功打开软件，一次性生图成功

sxx297015517

这个可以生成动图吗

YRGWZJ

大佬，5070ti16G运行图片编辑，只加载了一致性lora报错如下，找deep seek看了一下似乎是显存不足，请问有解决方法吗？谢谢！
[ComfyUI-Manager] Due to a network error, switching to local mode.
=> custom-node-list.json
=>
FETCH DATA from: D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\custom_nodes\ComfyUI-Manager\custom-node-list.json [DONE]
[INFO] [ComfyUI-Manager] All startup tasks have been completed.
[INFO] got prompt
[INFO] Requested to load Krea2TEModel_
[INFO] Model Krea2TEModel_ prepared for dynamic VRAM loading. 4999MB Staged. 0 patches attached. Force pre-loaded 249 weights: 627 KB.
[INFO] Model Krea2TEModel_ prepared for dynamic VRAM loading. 4999MB Staged. 0 patches attached. Force pre-loaded 249 weights: 627 KB.
[INFO] Requested to load WanVAE
[INFO] 0 models unloaded.
[INFO] Model WanVAE prepared for dynamic VRAM loading. 241MB Staged. 0 patches attached. Force pre-loaded 60 weights: 61 KB.
[rgthree-comfy][Power Lora Loader] Matched Lora input "图片编辑Lora模型/图片编辑专用krea2_identity_edit_v1_2.safetensors" to "图片编辑Lora模型\图片编辑专用krea2_identity_edit_v1_2.safetensors".
[INFO] Requested to load Krea2
[INFO] Model Krea2 prepared for dynamic VRAM loading. 12224MB Staged. 256 patches attached. Force pre-loaded 160 weights: 1431 KB.

  0%|          | 0/10 [00:00<?, ?it/s]
  0%|          | 0/10 [00:00<?, ?it/s,   Model Initializing ...  ][krea2edit] pixel path ACTIVE (fit_mode=fit)
[krea2edit] _fit_encode_image: mode=fit in=(1, 4096, 2304, 3) target_latent=512x288
[INFO] 0 models unloaded.
[krea2edit] STRIDE1-POS fit: ref grids [(256, 144)] centered in (256,144)

  0%|          | 0/10 [00:03<?, ?it/s,   Model Initializing ...  ]
[ERROR] !!! Exception during processing !!! Allocation on device
[ERROR] Traceback (most recent call last):
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\execution.py", line 542, 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 "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\execution.py", line 341, 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 "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\execution.py", line 315, in _async_map_node_over_list
    await process_inputs(input_dict, i)
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\execution.py", line 303, in process_inputs
    result = f(**inputs)
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\nodes.py", line 1584, in sample
    return common_ksampler(model, seed, steps, cfg, sampler_name, scheduler, positive, negative, latent_image, denoise=denoise)
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\nodes.py", line 1548, in common_ksampler
    samples = comfy.sample.sample(model, noise, steps, cfg, sampler_name, scheduler, positive, negative, latent_image,
                                  denoise=denoise, disable_noise=disable_noise, start_step=start_step, last_step=last_step,
                                  force_full_denoise=force_full_denoise, noise_mask=noise_mask, callback=callback, disable_pbar=disable_pbar, seed=seed)
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\comfy\sample.py", line 74, in sample
    samples = sampler.sample(noise, positive, negative, cfg=cfg, latent_image=latent_image, start_step=start_step, last_step=last_step, force_full_denoise=force_full_denoise, denoise_mask=noise_mask, sigmas=sigmas, callback=callback, disable_pbar=disable_pbar, seed=seed)
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\comfy\samplers.py", line 1444, in sample
    return sample(self.model, noise, positive, negative, cfg, self.device, sampler, sigmas, self.model_options, latent_image=latent_image, denoise_mask=denoise_mask, callback=callback, disable_pbar=disable_pbar, seed=seed)
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\comfy\samplers.py", line 1334, in sample
    return cfg_guider.sample(noise, latent_image, sampler, sigmas, denoise_mask, callback, disable_pbar, seed)
           ~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\comfy\samplers.py", line 1316, in sample
    output = executor.execute(noise, latent_image, sampler, sigmas, denoise_mask, callback, disable_pbar, seed, latent_shapes=latent_shapes)
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\comfy\patcher_extension.py", line 113, in execute
    return self.original(*args, **kwargs)
           ~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\comfy\samplers.py", line 1254, in outer_sample
    output = self.inner_sample(noise, latent_image, device, sampler, sigmas, denoise_mask, callback, disable_pbar, seed, latent_shapes=latent_shapes)
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\comfy\samplers.py", line 1229, in inner_sample
    samples = executor.execute(self, sigmas, extra_args, callback, noise, latent_image, denoise_mask, disable_pbar)
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\comfy\patcher_extension.py", line 113, in execute
    return self.original(*args, **kwargs)
           ~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\comfy\samplers.py", line 999, in sample
    samples = self.sampler_function(model_k, noise, sigmas, extra_args=extra_args, callback=k_callback, disable=disable_pbar, **self.extra_options)
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\python\Lib\site-packages\torch\utils\_contextlib.py", line 120, in decorate_context
    return func(*args, **kwargs)
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\comfy\k_diffusion\sampling.py", line 205, in sample_euler
    denoised = model(x, sigma_hat * s_in, **extra_args)
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\comfy\samplers.py", line 639, in __call__
    out = self.inner_model(x, sigma, model_options=model_options, seed=seed)
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\comfy\samplers.py", line 1202, in __call__
    return self.outer_predict_noise(*args, **kwargs)
           ~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\comfy\samplers.py", line 1209, in outer_predict_noise
    ).execute(x, timestep, model_options, seed)
      ~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\comfy\patcher_extension.py", line 113, in execute
    return self.original(*args, **kwargs)
           ~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\comfy\samplers.py", line 1212, in predict_noise
    return sampling_function(self.inner_model, x, timestep, self.conds.get("negative", None), self.conds.get("positive", None), self.cfg, model_options=model_options, seed=seed)
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\comfy\samplers.py", line 619, in sampling_function
    out = calc_cond_batch(model, conds, x, timestep, model_options)
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\comfy\samplers.py", line 210, in calc_cond_batch
    return _calc_cond_batch_outer(model, conds, x_in, timestep, model_options)
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\comfy\samplers.py", line 218, in _calc_cond_batch_outer
    return executor.execute(model, conds, x_in, timestep, model_options)
           ~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\comfy\patcher_extension.py", line 113, in execute
    return self.original(*args, **kwargs)
           ~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\comfy\samplers.py", line 334, in _calc_cond_batch
    output = model.apply_model(input_x, timestep_, **c).chunk(batch_chunks)
             ~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\comfy\model_base.py", line 191, in apply_model
    return comfy.patcher_extension.WrapperExecutor.new_class_executor(
           ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    ...<2 lines>...
        comfy.patcher_extension.get_all_wrappers(comfy.patcher_extension.WrappersMP.APPLY_MODEL, transformer_options)
        ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    ).execute(x, t, c_concat, c_crossattn, control, transformer_options, **kwargs)
    ~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\comfy\patcher_extension.py", line 113, in execute
    return self.original(*args, **kwargs)
           ~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\comfy\model_base.py", line 235, in _apply_model
    model_output = self.diffusion_model(xc, t, context=context, control=control, transformer_options=transformer_options, **extra_conds)
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\python\Lib\site-packages\torch\nn\modules\module.py", line 1775, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\python\Lib\site-packages\torch\nn\modules\module.py", line 1786, in _call_impl
    return forward_call(*args, **kwargs)
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\comfy\ldm\krea2\model.py", line 225, in forward
    return comfy.patcher_extension.WrapperExecutor.new_class_executor(
           ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    ...<2 lines>...
        comfy.patcher_extension.get_all_wrappers(comfy.patcher_extension.WrappersMP.DIFFUSION_MODEL, transformer_options),
        ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    ).execute(x, timesteps, context, attention_mask, transformer_options, **kwargs)
    ~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\comfy\patcher_extension.py", line 114, in execute
    return self.wrappers[self.idx](self, *args, **kwargs)
           ~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\custom_nodes\comfyui-krea2edit\__init__.py", line 321, in wrapper
    v = krea2_edit_forward(dm, x, timesteps, context, src, transformer_options,
                           ref_boost=ref_boost, ref_boost_a=ref_boost_a,
    ...<2 lines>...
                                       and source_image is not None),
                           pos_mode=("stride1" if fit_mode == "fit" else "anchor"))
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\custom_nodes\comfyui-krea2edit\__init__.py", line 237, in krea2_edit_forward
    combined = block(combined, tvec, freqs, attn_bias, transformer_options=transformer_options)
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\python\Lib\site-packages\torch\nn\modules\module.py", line 1775, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\python\Lib\site-packages\torch\nn\modules\module.py", line 1786, in _call_impl
    return forward_call(*args, **kwargs)
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\comfy\ldm\krea2\model.py", line 163, in forward
    x = x + pregate * self.attn((1 + prescale) * self.prenorm(x) + preshift, freqs, mask, transformer_options=transformer_options)
                                                 ~~~~~~~~~~~~^^^
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\python\Lib\site-packages\torch\nn\modules\module.py", line 1775, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\python\Lib\site-packages\torch\nn\modules\module.py", line 1786, in _call_impl
    return forward_call(*args, **kwargs)
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\ComfyUI\comfy\ldm\krea2\model.py", line 34, in forward
    return F.rms_norm(x.float(), (x.shape[-1],), weight=weight, eps=self.eps).to(dtype)
           ~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\JianxiaAI-CreativePack-v2.2\JianxiaAI-CreativePack-v2.2\python\Lib\site-packages\torch\nn\functional.py", line 2920, in rms_norm
    return torch.rms_norm(input, normalized_shape, weight, eps)
           ~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
torch.OutOfMemoryError: Allocation on device

妫盎司

如果启动超时，可以开一个github的加速器

sxx297015517

大佬能看下怎么回事吗？3060的卡
[launcher] 启动 ComfyUI
[INFO] setup plugin alembic.autogenerate.schemas
[INFO] setup plugin alembic.autogenerate.tables
[INFO] setup plugin alembic.autogenerate.types
[INFO] setup plugin alembic.autogenerate.constraints
[INFO] setup plugin alembic.autogenerate.defaults
[INFO] setup plugin alembic.autogenerate.comments
[INFO] Setting output directory to: C:\AI\output
[ERROR] Failed to execute startup-script: C:\AI\ComfyUI\custom_nodes\ComfyUI-Manager\prestartup_script.py / Failed to initialize: Bad git executable.
The git executable must be specified in one of the following ways:
    - be included in your $PATH
    - be set via $GIT_PYTHON_GIT_EXECUTABLE
    - explicitly set via git.refresh(<full-path-to-git-executable>)

All git commands will error until this is rectified.

This initial message can be silenced or aggravated in the future by setting the
$GIT_PYTHON_REFRESH environment variable. Use one of the following values:
    - quiet|q|silence|s|silent|none|n|0: for no message or exception
    - warn|w|warning|log|l|1: for a warning message (logging level CRITICAL, displayed by default)
    - error|e|exception|raise|r|2: for a raised exception

Example:
    export GIT_PYTHON_REFRESH=quiet

[INFO]
Prestartup times for custom nodes:
[INFO]    0.0 seconds: C:\AI\ComfyUI\custom_nodes\rgthree-comfy
[INFO]    0.0 seconds: C:\AI\ComfyUI\custom_nodes\comfyui-easy-use
[INFO]    0.0 seconds: C:\AI\ComfyUI\custom_nodes\deno-custom-nodes
[INFO]    1.8 seconds (PRESTARTUP FAILED): C:\AI\ComfyUI\custom_nodes\ComfyUI-Manager
[INFO]
C:\AI\python\Lib\site-packages\torch\cuda\__init__.py:182: UserWarning: cudaGetDeviceCount() returned cudaErrorNotSupported, likely using older driver or on CPU machine (Triggered internally at C:\actions-runner\_work\pytorch\pytorch\pytorch\c10\cuda\CUDAFunctions.cpp:88.)
  return torch._C._cuda_getDeviceCount() > 0
[INFO] Found comfy_kitchen backend cuda: {'available': False, 'disabled': False, 'unavailable_reason': 'CUDA not available on this system', 'capabilities': []}
[INFO] Found comfy_kitchen backend triton: {'available': False, 'disabled': True, 'unavailable_reason': 'Neither CUDA nor XPU available on this system', 'capabilities': []}
[INFO] Found comfy_kitchen backend eager: {'available': True, 'disabled': False, 'unavailable_reason': None, 'capabilities': ['adaln', 'apply_rope', 'apply_rope1', 'apply_rope_split_half', 'apply_rope_split_half1', 'convrot_w4a4_linear', 'dequantize_convrot_w4a4_weight', 'dequantize_int8_convrot_weight', 'dequantize_int8_convrot_weight_dtype', 'dequantize_int8_simple', 'dequantize_int8_simple_dtype', 'dequantize_mxfp8', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'gemv_awq_w4a16', 'int8_linear', 'prepare_int4_weight_for_int8_linear', 'quantize_and_rotate_rowwise', 'quantize_convrot_w4a4_weight', 'quantize_int8_convrot_weight', 'quantize_int8_rowwise', 'quantize_int8_tensorwise', 'quantize_mxfp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8', 'quantize_svdquant_w4a4', 'rms_rope', 'rms_rope1', 'rms_rope_split_half', 'rms_rope_split_half1', 'scaled_mm_mxfp8', 'scaled_mm_nvfp4', 'scaled_mm_svdquant_w4a4', 'stochastic_rounding_fp8']}
[INFO] Checkpoint files will always be loaded safely.
Windows fatal exception: access violation

Stack (most recent call first):
  File "C:\AI\python\Lib\site-packages\torch\cuda\__init__.py", line 410 in _lazy_init
  File "C:\AI\python\Lib\site-packages\torch\cuda\__init__.py", line 1069 in current_device
  File "C:\AI\ComfyUI\comfy\model_management.py", line 211 in get_torch_device
  File "C:\AI\ComfyUI\comfy\model_management.py", line 362 in <module>
  File "<frozen importlib._bootstrap>", line 488 in _call_with_frames_removed
  File "<frozen importlib._bootstrap_external>", line 1023 in exec_module
  File "<frozen importlib._bootstrap>", line 935 in _load_unlocked
  File "<frozen importlib._bootstrap>", line 1331 in _find_and_load_unlocked
  File "<frozen importlib._bootstrap>", line 1360 in _find_and_load
  File "C:\AI\ComfyUI\execution.py", line 18 in <module>
  File "<frozen importlib._bootstrap>", line 488 in _call_with_frames_removed
  File "<frozen importlib._bootstrap_external>", line 1023 in exec_module
  File "<frozen importlib._bootstrap>", line 935 in _load_unlocked
  File "<frozen importlib._bootstrap>", line 1331 in _find_and_load_unlocked
  File "<frozen importlib._bootstrap>", line 1360 in _find_and_load
  File "C:\AI\ComfyUI\main.py", line 227 in <module>

[release] 启动失败
Traceback (most recent call last):
  File "release.py", line 52, in run_release
RuntimeError: ComfyUI 启动超时，请查看 data/runtime.log

sxx297015517

大佬能给看看怎么回事吗，3060的卡，12G，本地监测都通过了
[launcher] 启动 ComfyUI
[INFO] setup plugin alembic.autogenerate.schemas
[INFO] setup plugin alembic.autogenerate.tables
[INFO] setup plugin alembic.autogenerate.types
[INFO] setup plugin alembic.autogenerate.constraints
[INFO] setup plugin alembic.autogenerate.defaults
[INFO] setup plugin alembic.autogenerate.comments
[INFO] Setting output directory to: C:\AI\output
[ERROR] Failed to execute startup-script: C:\AI\ComfyUI\custom_nodes\ComfyUI-Manager\prestartup_script.py / Failed to initialize: Bad git executable.
The git executable must be specified in one of the following ways:
    - be included in your $PATH
    - be set via $GIT_PYTHON_GIT_EXECUTABLE
    - explicitly set via git.refresh(<full-path-to-git-executable>)

All git commands will error until this is rectified.

This initial message can be silenced or aggravated in the future by setting the
$GIT_PYTHON_REFRESH environment variable. Use one of the following values:
    - quiet|q|silence|s|silent|none|n|0: for no message or exception
    - warn|w|warning|log|l|1: for a warning message (logging level CRITICAL, displayed by default)
    - error|e|exception|raise|r|2: for a raised exception

Example:
    export GIT_PYTHON_REFRESH=quiet

[INFO]
Prestartup times for custom nodes:
[INFO]    0.0 seconds: C:\AI\ComfyUI\custom_nodes\rgthree-comfy
[INFO]    0.0 seconds: C:\AI\ComfyUI\custom_nodes\comfyui-easy-use
[INFO]    0.0 seconds: C:\AI\ComfyUI\custom_nodes\deno-custom-nodes
[INFO]    1.8 seconds (PRESTARTUP FAILED): C:\AI\ComfyUI\custom_nodes\ComfyUI-Manager
[INFO]
C:\AI\python\Lib\site-packages\torch\cuda\__init__.py:182: UserWarning: cudaGetDeviceCount() returned cudaErrorNotSupported, likely using older driver or on CPU machine (Triggered internally at C:\actions-runner\_work\pytorch\pytorch\pytorch\c10\cuda\CUDAFunctions.cpp:88.)
  return torch._C._cuda_getDeviceCount() > 0
[INFO] Found comfy_kitchen backend cuda: {'available': False, 'disabled': False, 'unavailable_reason': 'CUDA not available on this system', 'capabilities': []}
[INFO] Found comfy_kitchen backend triton: {'available': False, 'disabled': True, 'unavailable_reason': 'Neither CUDA nor XPU available on this system', 'capabilities': []}
[INFO] Found comfy_kitchen backend eager: {'available': True, 'disabled': False, 'unavailable_reason': None, 'capabilities': ['adaln', 'apply_rope', 'apply_rope1', 'apply_rope_split_half', 'apply_rope_split_half1', 'convrot_w4a4_linear', 'dequantize_convrot_w4a4_weight', 'dequantize_int8_convrot_weight', 'dequantize_int8_convrot_weight_dtype', 'dequantize_int8_simple', 'dequantize_int8_simple_dtype', 'dequantize_mxfp8', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'gemv_awq_w4a16', 'int8_linear', 'prepare_int4_weight_for_int8_linear', 'quantize_and_rotate_rowwise', 'quantize_convrot_w4a4_weight', 'quantize_int8_convrot_weight', 'quantize_int8_rowwise', 'quantize_int8_tensorwise', 'quantize_mxfp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8', 'quantize_svdquant_w4a4', 'rms_rope', 'rms_rope1', 'rms_rope_split_half', 'rms_rope_split_half1', 'scaled_mm_mxfp8', 'scaled_mm_nvfp4', 'scaled_mm_svdquant_w4a4', 'stochastic_rounding_fp8']}
[INFO] Checkpoint files will always be loaded safely.
Windows fatal exception: access violation

Stack (most recent call first):
  File "C:\AI\python\Lib\site-packages\torch\cuda\__init__.py", line 410 in _lazy_init
  File "C:\AI\python\Lib\site-packages\torch\cuda\__init__.py", line 1069 in current_device
  File "C:\AI\ComfyUI\comfy\model_management.py", line 211 in get_torch_device
  File "C:\AI\ComfyUI\comfy\model_management.py", line 362 in <module>
  File "<frozen importlib._bootstrap>", line 488 in _call_with_frames_removed
  File "<frozen importlib._bootstrap_external>", line 1023 in exec_module
  File "<frozen importlib._bootstrap>", line 935 in _load_unlocked
  File "<frozen importlib._bootstrap>", line 1331 in _find_and_load_unlocked
  File "<frozen importlib._bootstrap>", line 1360 in _find_and_load
  File "C:\AI\ComfyUI\execution.py", line 18 in <module>
  File "<frozen importlib._bootstrap>", line 488 in _call_with_frames_removed
  File "<frozen importlib._bootstrap_external>", line 1023 in exec_module
  File "<frozen importlib._bootstrap>", line 935 in _load_unlocked
  File "<frozen importlib._bootstrap>", line 1331 in _find_and_load_unlocked
  File "<frozen importlib._bootstrap>", line 1360 in _find_and_load
  File "C:\AI\ComfyUI\main.py", line 227 in <module>

[release] 启动失败
Traceback (most recent call last):
  File "release.py", line 52, in run_release
RuntimeError: ComfyUI 启动超时，请查看 data/runtime.log

性爱之神

大佬，我只想要个去衣的，有没有？图片一键去衣或者批量图片去衣，或者换装也行，就是能够把其他女人的裸体直接换上去的。或者视频把其他女人的裸体换上去的。

yuanmouren007

qwen3vl_4b_fp8_scaled.safetensors这个文件在哪

静若

sxx297015517 发表于 2026-7-31 07:23
大佬能给看看怎么回事吗，3060的卡，12G，本地监测都通过了
[launcher] 启动 ComfyUI
 setup plugin ...

你的显卡驱动太旧了，CUDA最低需要13.3，更新显卡驱动到最新，然后重启电脑在试试

静若

yuanmouren007 发表于 2026-7-31 10:27
qwen3vl_4b_fp8_scaled.safetensors这个文件在哪

在整合包根目录，ComfyUI / models / clip

sxx297015517

静若 发表于 2026-7-31 15:05
你的显卡驱动太旧了，CUDA最低需要13.3，更新显卡驱动到最新，然后重启电脑在试试
...

更新后显示版本530，还是不行，有什么办法吗

静若

sxx297015517 发表于 2026-7-31 18:39
更新后显示版本530，还是不行，有什么办法吗

在桌面按下 Win + R 弹出【运行】窗口输入：cmd，回车，在终端输入：nvidia-smi，即可查看，CUDAxx对应的版本号，如果显示低于CUDA UMD Version: 13.3，那就无法兼容整合包，会直接启动失败，如果显示；CUDA UMD Version: 13.3，那说明是其他问题，可能是没有安装； VC++ 运行库

yuanmouren007

如何保持面部不变呢，教教我

sxx297015517

静若 发表于 2026-7-31 22:58
在桌面按下 Win + R 弹出【运行】窗口输入：cmd，回车，在终端输入：nvidia-smi，即可查看，CUDAxx对应的 ...

可以了，感谢大佬，如果要自己加lora需要用什么格式的呀