RuntimeError: shape '[1, 68, 120, 16, 2, 2]' is invalid for input of size 1044480

### Custom Node Testing

- [ ] I have tried disabling custom nodes and the issue persists (see [how to disable custom nodes](https://docs.comfy.org/troubleshooting/custom-node-issues#step-1%3A-test-with-all-custom-nodes-disabled) if you need help)

### Expected Behavior

无法正常运行

### Actual Behavior

RuntimeError: shape '[1, 68, 120, 16, 2, 2]' is invalid for input of size 1044480

### Steps to Reproduce

无

### Debug Logs

```powershell
RuntimeError: shape '[1, 68, 120, 16, 2, 2]' is invalid for input of size 1044480

# ComfyUI Error Report
## Error Details
- **Node ID:** 258
- **Node Type:** KSampler (Efficient)
- **Exception Type:** RuntimeError
- **Exception Message:** RuntimeError: shape '[1, 68, 120, 16, 2, 2]' is invalid for input of size 1044480

## Stack Trace

File "D:\AI\ComfyUI\execution.py", line 536, 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:\AI\ComfyUI\execution.py", line 336, 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:\AI\ComfyUI\execution.py", line 310, in _async_map_node_over_list
await process_inputs(input_dict, i)

File "D:\AI\ComfyUI\execution.py", line 298, in process_inputs
result = f(**inputs)

File "D:\AI\ComfyUI\custom_nodes\efficiency-nodes-comfyui\efficiency_nodes.py", line 741, in sample
samples, images, gifs, preview = process_latent_image(model, seed, steps, cfg, sampler_name, scheduler,
~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
positive, negative, latent_image, denoise, sampler_type, add_noise,
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
start_at_step, end_at_step, return_with_leftover_noise, refiner_model,
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
refiner_positive, refiner_negative, vae, vae_decode, preview_method)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "D:\AI\ComfyUI\custom_nodes\efficiency-nodes-comfyui\efficiency_nodes.py", line 559, in process_latent_image
samples = KSampler().sample(model, seed, steps, cfg, sampler_name, scheduler, positive, negative,
~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
latent_image, denoise=denoise)[0] if denoise>0 else latent_image
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "D:\AI\ComfyUI\nodes.py", line 1586, in sample
return common_ksampler(model, seed, steps, cfg, sampler_name, scheduler, positive, negative, latent_image, denoise=denoise)

File "D:\AI\ComfyUI\nodes.py", line 1550, 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:\AI\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:\AI\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:\AI\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:\AI\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:\AI\ComfyUI\comfy\patcher_extension.py", line 113, in execute
return self.original(*args, **kwargs)
~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^

File "D:\AI\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:\AI\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:\AI\ComfyUI\comfy\patcher_extension.py", line 113, in execute
return self.original(*args, **kwargs)
~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^

File "D:\AI\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:\AI\python-3.13.13\comfyui\Lib\site-packages\torch\utils\_contextlib.py", line 120, in decorate_context
return func(*args, **kwargs)

File "D:\AI\ComfyUI\comfy\k_diffusion\sampling.py", line 205, in sample_euler
denoised = model(x, sigma_hat * s_in, **extra_args)

File "D:\AI\ComfyUI\comfy\samplers.py", line 639, in __call__
out = self.inner_model(x, sigma, model_options=model_options, seed=seed)

File "D:\AI\ComfyUI\comfy\samplers.py", line 1202, in __call__
return self.outer_predict_noise(*args, **kwargs)
~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^

File "D:\AI\ComfyUI\comfy\samplers.py", line 1209, in outer_predict_noise
).execute(x, timestep, model_options, seed)
~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "D:\AI\ComfyUI\comfy\patcher_extension.py", line 113, in execute
return self.original(*args, **kwargs)
~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^

File "D:\AI\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:\AI\ComfyUI\comfy\samplers.py", line 619, in sampling_function
out = calc_cond_batch(model, conds, x, timestep, model_options)

File "D:\AI\ComfyUI\comfy\samplers.py", line 210, in calc_cond_batch
return _calc_cond_batch_outer(model, conds, x_in, timestep, model_options)

File "D:\AI\ComfyUI\comfy\samplers.py", line 218, in _calc_cond_batch_outer
return executor.execute(model, conds, x_in, timestep, model_options)
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "D:\AI\ComfyUI\comfy\patcher_extension.py", line 113, in execute
return self.original(*args, **kwargs)
~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^

File "D:\AI\ComfyUI\comfy\samplers.py", line 334, in _calc_cond_batch
output = model.apply_model(input_x, timestep_, **c).chunk(batch_chunks)
~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^

File "D:\AI\ComfyUI\comfy\model_base.py", line 187, 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:\AI\ComfyUI\comfy\patcher_extension.py", line 113, in execute
return self.original(*args, **kwargs)
~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^

File "D:\AI\ComfyUI\comfy\model_base.py", line 231, in _apply_model
model_output = self.diffusion_model(xc, t, context=context, control=control, transformer_options=transformer_options, **extra_conds)

File "D:\AI\python-3.13.13\comfyui\Lib\site-packages\torch\nn\modules\module.py", line 1775, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^

File "D:\AI\python-3.13.13\comfyui\Lib\site-packages\torch\nn\modules\module.py", line 1786, in _call_impl
return forward_call(*args, **kwargs)

File "D:\AI\ComfyUI\custom_nodes\ComfyUI-QwenImageLoraLoader\wrappers\qwenimage.py", line 295, in forward
out = self._execute_model(x, timestep, context, guidance, control, transformer_options, **kwargs)

File "D:\AI\ComfyUI\custom_nodes\ComfyUI-QwenImageLoraLoader\wrappers\qwenimage.py", line 414, in _execute_model
return self.model(
~~~~~~~~~~^
x,
^^
...<6 lines>...
**final_kwargs_cleaned,
^^^^^^^^^^^^^^^^^^^^^^^
)
^

File "D:\AI\python-3.13.13\comfyui\Lib\site-packages\torch\nn\modules\module.py", line 1775, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^

File "D:\AI\python-3.13.13\comfyui\Lib\site-packages\torch\nn\modules\module.py", line 1786, in _call_impl
return forward_call(*args, **kwargs)

File "D:\AI\ComfyUI\comfy\ldm\qwen_image\model.py", line 419, 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, timestep, context, attention_mask, ref_latents, additional_t_cond, transformer_options, **kwargs)
~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "D:\AI\ComfyUI\comfy\patcher_extension.py", line 113, in execute
return self.original(*args, **kwargs)
~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^

File "D:\AI\ComfyUI\custom_nodes\ComfyUI-nunchaku\models\qwenimage.py", line 813, in _forward
hidden_states = hidden_states[:, :num_embeds].view(
orig_shape[0], orig_shape[-2] // 2, orig_shape[-1] // 2, orig_shape[1], 2, 2
)

## System Information
- **ComfyUI Version:** 0.24.0
- **Arguments:** D:\AI\ComfyUI\main.py --windows-standalone-build --fast fp16_accumulation --enable-manager
- **OS:** win32
- **Python Version:** 3.13.13 (tags/v3.13.13:01104ce, Apr 7 2026, 19:25:48) [MSC v.1944 64 bit (AMD64)]
- **Embedded Python:** false
- **PyTorch Version:** 2.9.0+cu130
## Devices

- **Name:** cuda:0 NVIDIA GeForce RTX 4070 : cudaMallocAsync
- **Type:** cuda
- **VRAM Total:** 12878086144
- **VRAM Free:** 11605508096
- **Torch VRAM Total:** 33554432
- **Torch VRAM Free:** 25034752

## Logs

2026-06-07T10:56:20.107389 -
2026-06-07T10:56:20.107644 - [FlashVSR] Starting VAE decoding...2026-06-07T10:56:20.107855 -
2026-06-07T10:56:21.941633 - [FlashVSR] Processing tile 7/15: coords (232,232) to (488,488)2026-06-07T10:56:21.941802 -
2026-06-07T10:56:26.141331 -
100%|████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:01<00:00, 1.02s/it]2026-06-07T10:56:26.141408 -
2026-06-07T10:56:26.141583 - [FlashVSR] Starting VAE decoding...2026-06-07T10:56:26.141676 -
2026-06-07T10:56:27.924922 - [FlashVSR] Processing tile 8/15: coords (464,232) to (720,488)2026-06-07T10:56:27.925083 -
2026-06-07T10:56:32.143752 -
100%|████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:01<00:00, 1.02s/it]2026-06-07T10:56:32.143839 -
2026-06-07T10:56:32.144054 - [FlashVSR] Starting VAE decoding...2026-06-07T10:56:32.144112 -
2026-06-07T10:56:33.932764 - [FlashVSR] Processing tile 9/15: coords (696,232) to (952,488)2026-06-07T10:56:33.933196 -
2026-06-07T10:56:38.086688 -
100%|████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:01<00:00, 1.01s/it]2026-06-07T10:56:38.086801 -
2026-06-07T10:56:38.087014 - [FlashVSR] Starting VAE decoding...2026-06-07T10:56:38.087105 -
2026-06-07T10:56:39.926351 - [FlashVSR] Processing tile 10/15: coords (768,232) to (1024,488)2026-06-07T10:56:39.926526 -
2026-06-07T10:56:46.802067 -
100%|████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:01<00:00, 1.03s/it]2026-06-07T10:56:46.802208 -
2026-06-07T10:56:46.802531 - [FlashVSR] Starting VAE decoding...2026-06-07T10:56:46.802660 -
2026-06-07T10:56:48.605008 - [FlashVSR] Processing tile 11/15: coords (0,324) to (256,580)2026-06-07T10:56:48.605157 -
2026-06-07T10:56:53.010214 -
100%|████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:00<00:00, 1.01it/s]2026-06-07T10:56:53.010331 -
2026-06-07T10:56:53.010526 - [FlashVSR] Starting VAE decoding...2026-06-07T10:56:53.010614 -
2026-06-07T10:56:54.851845 - [FlashVSR] Processing tile 12/15: coords (232,324) to (488,580)2026-06-07T10:56:54.852027 -
2026-06-07T10:56:58.909838 -
100%|████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:00<00:00, 1.05it/s]2026-06-07T10:56:58.909941 -
2026-06-07T10:56:58.910169 - [FlashVSR] Starting VAE decoding...2026-06-07T10:56:58.910253 -
2026-06-07T10:57:00.709451 - [FlashVSR] Processing tile 13/15: coords (464,324) to (720,580)2026-06-07T10:57:00.709611 -
2026-06-07T10:57:04.709170 -
100%|████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:00<00:00, 1.05it/s]2026-06-07T10:57:04.709233 -
2026-06-07T10:57:04.709439 - [FlashVSR] Starting VAE decoding...2026-06-07T10:57:04.709497 -
2026-06-07T10:57:06.435974 - [FlashVSR] Processing tile 14/15: coords (696,324) to (952,580)2026-06-07T10:57:06.436107 -
2026-06-07T10:57:10.472386 -
100%|████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:00<00:00, 1.03it/s]2026-06-07T10:57:10.472490 -
2026-06-07T10:57:10.472726 - [FlashVSR] Starting VAE decoding...2026-06-07T10:57:10.472943 -
2026-06-07T10:57:12.210986 - [FlashVSR] Processing tile 15/15: coords (768,324) to (1024,580)2026-06-07T10:57:12.211148 -
2026-06-07T10:57:18.865610 -
100%|████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:03<00:00, 3.66s/it]2026-06-07T10:57:18.865721 -
2026-06-07T10:57:18.865934 - [FlashVSR] Starting VAE decoding...2026-06-07T10:57:18.866010 -
2026-06-07T10:57:20.721636 - [FlashVSR] Done.2026-06-07T10:57:20.721890 -
2026-06-07T10:57:21.650772 - [INFO] Prompt executed in 121.26 seconds
2026-06-07T11:03:43.084358 - [INFO] Using RAM pressure cache.
2026-06-07T11:03:51.714925 - [INFO] got prompt
2026-06-07T11:03:52.362890 - 2026-06-07T11:03:52.368514 -
2026-06-07T11:03:52.373504 - 2026-06-07T11:03:52.379403 -
2026-06-07T11:03:52.379518 - ███████╗███████╗███████╗██████╗ ██╗ ██╗██████╗ ██████╗ ███████╗2026-06-07T11:03:52.385390 -
2026-06-07T11:03:52.396401 - ██╔════╝██╔════╝██╔════╝██╔══██╗██║ ██║██╔══██╗ ╚════██╗ ██╔════╝2026-06-07T11:03:52.406405 -
2026-06-07T11:03:52.417386 - ███████╗█████╗ █████╗ ██║ ██║██║ ██║██████╔╝ █████╔╝ ███████╗2026-06-07T11:03:52.428402 -
2026-06-07T11:03:52.440435 - ╚════██║██╔══╝ ██╔══╝ ██║ ██║╚██╗ ██╔╝██╔══██╗ ██╔═══╝ ╚════██║2026-06-07T11:03:52.451142 -
2026-06-07T11:03:52.456414 - ███████║███████╗███████╗██████╔╝ ╚████╔╝ ██║ ██║ ███████╗ ██╗ ███████║2026-06-07T11:03:52.472734 -
2026-06-07T11:03:52.484253 - ╚══════╝╚══════╝╚══════╝╚═════╝ ╚═══╝ ╚═╝ ╚═╝ ╚══════╝ ╚═╝ ╚══════╝2026-06-07T11:03:52.496250 -
2026-06-07T11:03:52.507254 - v2.5.24 © ByteDance Seed · NumZ · AInVFX2026-06-07T11:03:52.518256 -
2026-06-07T11:03:52.529332 - ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━2026-06-07T11:03:52.542275 -
2026-06-07T11:03:52.542489 - 2026-06-07T11:03:52.553274 -
2026-06-07T11:03:52.740555 - [11:03:52.735] ♻️ Reusing cached runner template: nodes 3+42026-06-07T11:03:52.751466 -
2026-06-07T11:03:52.761990 - [11:03:52.756] ♻️ Reusing cached DiT (3): seedvr2_ema_7b-Q8_0.gguf2026-06-07T11:03:52.773086 -
2026-06-07T11:03:52.862235 - [11:03:52.857] ♻️ Reusing cached VAE (4): ema_vae_fp16.safetensors2026-06-07T11:03:52.873230 -
2026-06-07T11:03:53.786383 - [11:03:53.781] 2026-06-07T11:03:53.797452 -
2026-06-07T11:03:53.807470 - [11:03:53.802] 🎬 Starting upscaling generation...2026-06-07T11:03:53.818442 -
2026-06-07T11:03:53.829444 - [11:03:53.824] 🎬 Input: 1 frame, 1536x870px → Output: 4096x2320px (shortest edge: 4096px, max edge: 4096px)2026-06-07T11:03:53.840009 -
2026-06-07T11:03:53.845952 - [11:03:53.845] 🎬 Batch size: 1, Seed: 649857867, Channels: RGB2026-06-07T11:03:53.850714 -
2026-06-07T11:03:53.850793 - [11:03:53.850] 2026-06-07T11:03:53.850834 -
2026-06-07T11:03:53.862159 - [11:03:53.856] ━━━━━━━━ Phase 1: VAE encoding ━━━━━━━━2026-06-07T11:03:53.872745 -
2026-06-07T11:03:56.321750 - [11:03:56.316] 🎨 Encoding batch 1/12026-06-07T11:03:56.333285 -
2026-06-07T11:03:56.363374 - [11:03:56.357] 📹 Sequence of 1 frames2026-06-07T11:03:56.373436 -
2026-06-07T11:03:56.772720 - [11:03:56.767] 🎨 Using VAE tiled encoding (Tile: (1024, 1024), Overlap: (128, 128))2026-06-07T11:03:56.782731 -
2026-06-07T11:04:03.518545 - [11:04:03.518] 2026-06-07T11:04:03.518794 -
2026-06-07T11:04:03.519056 - [11:04:03.519] ━━━━━━━━ Phase 2: DiT upscaling ━━━━━━━━2026-06-07T11:04:03.519198 -
2026-06-07T11:04:03.547038 - [11:04:03.546] 🎬 Upscaling batch 1/12026-06-07T11:04:03.547234 -
2026-06-07T11:04:31.799353 -
EulerSampler: 100%|██████████████████████████████████████████████████████████████████████| 1/1 [00:28<00:00, 28.25s/it]2026-06-07T11:04:31.799530 -
2026-06-07T11:04:31.805949 - [11:04:31.805] 2026-06-07T11:04:31.806240 -
2026-06-07T11:04:31.806509 - [11:04:31.806] ━━━━━━━━ Phase 3: VAE decoding ━━━━━━━━2026-06-07T11:04:31.806679 -
2026-06-07T11:04:31.806903 - [11:04:31.806] 🔧 Pre-allocating output tensor: 1 frames, 4096x2320px, RGB (0.05GB)2026-06-07T11:04:31.807020 -
2026-06-07T11:04:31.966239 - [11:04:31.966] 🎨 Decoding batch 1/12026-06-07T11:04:31.966362 -
2026-06-07T11:04:31.967575 - [11:04:31.967] 🎨 Using VAE tiled decoding (Tile: (1024, 1024), Overlap: (128, 128))2026-06-07T11:04:31.967717 -
2026-06-07T11:04:47.124269 - [11:04:47.124] 2026-06-07T11:04:47.124536 -
2026-06-07T11:04:47.124738 - [11:04:47.124] ━━━━━━━━ Phase 4: Post-processing ━━━━━━━━2026-06-07T11:04:47.124871 -
2026-06-07T11:04:47.125026 - [11:04:47.124] 📹 Post-processing batch 1/12026-06-07T11:04:47.125136 -
2026-06-07T11:04:47.400601 - [11:04:47.400] 📹 Applying LAB perceptual color transfer2026-06-07T11:04:47.400743 -
2026-06-07T11:04:47.544035 - [11:04:47.543] 🎬 Output assembled: 1 frames, Resolution: 4096x2320px, Channels: RGB2026-06-07T11:04:47.544251 -
2026-06-07T11:04:47.544844 - [11:04:47.544] 2026-06-07T11:04:47.544971 -
2026-06-07T11:04:47.555546 - [11:04:47.555] ✅ Upscaling completed successfully!2026-06-07T11:04:47.555970 -
2026-06-07T11:04:48.140741 - [11:04:48.140] 💾 Models cached for next run: DiT (seedvr2_ema_7b-Q8_0.gguf) and VAE (ema_vae_fp16.safetensors)2026-06-07T11:04:48.140945 -
2026-06-07T11:04:48.144384 - [11:04:48.144] ⚡ Average FPS: 0.02 frames/sec2026-06-07T11:04:48.144469 -
2026-06-07T11:04:48.144563 - [11:04:48.144] 2026-06-07T11:04:48.144625 -
2026-06-07T11:04:48.144711 - [11:04:48.144] ────────────────────────2026-06-07T11:04:48.144790 -
2026-06-07T11:04:48.144857 - [11:04:48.144] 💬 Questions? Updates? Watch, star & sponsor if you can!2026-06-07T11:04:48.144942 -
2026-06-07T11:04:48.145030 - [11:04:48.144] 🎬 https://www.youtube.com/@AInVFX2026-06-07T11:04:48.145106 -
2026-06-07T11:04:48.145168 - [11:04:48.145] ⭐💝 https://github.com/numz/ComfyUI-SeedVR2_VideoUpscaler2026-06-07T11:04:48.145217 -
2026-06-07T11:04:49.097816 - [INFO] Prompt executed in 57.38 seconds
2026-06-07T11:55:25.631623 - [INFO] Using RAM pressure cache.
2026-06-07T13:06:54.713552 - [INFO] Using RAM pressure cache.
2026-06-07T13:07:14.882326 - [INFO] got prompt
2026-06-07T13:07:14.913673 - [INFO] Using pytorch attention in VAE
2026-06-07T13:07:14.914990 - [INFO] Using pytorch attention in VAE
2026-06-07T13:07:16.902943 - [INFO] VAE load device: cuda:0, offload device: cpu, dtype: torch.bfloat16
2026-06-07T13:07:18.203348 - [INFO] Found quantization metadata version 1
2026-06-07T13:07:18.204556 - [INFO] Using MixedPrecisionOps for text encoder
2026-06-07T13:07:21.917750 - [INFO] CLIP/text encoder model load device: cuda:0, offload device: cpu, current: cpu, dtype: torch.float16
2026-06-07T13:07:21.924289 - [INFO] Requested to load QwenImageTEModel_
2026-06-07T13:07:22.035974 - [INFO] Model QwenImageTEModel_ prepared for dynamic VRAM loading. 4906MB Staged. 0 patches attached. Force pre-loaded 122 weights: 561 KB.
2026-06-07T13:07:24.026319 - [INFO] Model QwenImageTEModel_ prepared for dynamic VRAM loading. 4906MB Staged. 0 patches attached. Force pre-loaded 122 weights: 561 KB.
2026-06-07T13:07:24.628436 - [INFO] VRAM < 15GiB, enabling CPU offload
2026-06-07T13:07:24.631572 - [WARNING]
WARNING, you accessed scaled_fp8 from the model config object which doesn't exist. Please fix your code.

2026-06-07T13:07:24.735375 - [INFO] model weight dtype torch.bfloat16, manual cast: None
2026-06-07T13:07:24.751621 - [INFO] model_type FLUX
2026-06-07T13:07:39.892089 - [INFO] [LoRA Stack Status] toggle_all: True
2026-06-07T13:07:39.894987 - [INFO] [LoRA Stack Status] Processing 3 LoRA slot(s):
2026-06-07T13:07:39.895600 - [INFO] [LoRA Stack Status] Slot 1: | 'Qwen-Image-2512-Lightning\Qwen-Image-2512-Lightning-4steps-V1.0-bf16.safetensors' | strength=1.0 | toggle_all=True | enabled_1=True | final_enabled=True | → APPLIED ✓
2026-06-07T13:07:39.897296 - [INFO] [LoRA Stack Status] Slot 2: | 'Qwen-360-diffusion-2512-int8-bf16-v2.safetensors' | strength=1.0 | toggle_all=True | enabled_2=True | final_enabled=True | → APPLIED ✓
2026-06-07T13:07:39.897693 - [INFO] [LoRA Stack Status] Slot 3: | 'Qwen-image2512-anime2026.6.3-1_c1-st7000.safetensors' | strength=0.7 | toggle_all=True | enabled_3=True | final_enabled=True | → APPLIED ✓
2026-06-07T13:07:39.898953 - [INFO] [LoRA Stack Status] Summary: 3 LoRA(s) will be applied out of 3 slot(s)
2026-06-07T13:07:40.379721 - [INFO] 🔍 Model wrapper type: 'NunchakuQwenImageTransformer2DModel'
2026-06-07T13:07:40.380141 - [INFO] 🔍 Model wrapper module: D:\AI\ComfyUI\custom_nodes\ComfyUI-nunchaku.models.qwenimage
2026-06-07T13:07:40.380397 - [INFO] 🔍 Type repr:
2026-06-07T13:07:40.380610 - [INFO] 🔍 Has 'model' attr? False
2026-06-07T13:07:40.380718 - [INFO] 🔍 Has 'loras' attr? False
2026-06-07T13:07:40.381090 - [INFO] 🔧 Wrapping NunchakuQwenImageTransformer2DModel with ComfyQwenImageWrapper
2026-06-07T13:07:40.381309 - [INFO] 📦 Creating ComfyQwenImageWrapper with cpu_offload='disable', apply_awq_mod=True (V3: always enabled)
2026-06-07T13:07:40.383988 - [INFO] 🔧 CPU offload setting: 'disable' (VRAM margin: 4.0GB)
2026-06-07T13:07:40.393666 - [INFO] Total LoRAs in stack: 3
2026-06-07T13:07:40.427965 - [INFO] Using sage attention mode: auto
2026-06-07T13:07:40.547331 - [INFO] Requested to load NunchakuQwenImage
2026-06-07T13:07:40.616211 - [INFO] Patching torch settings: torch.backends.cuda.matmul.allow_fp16_accumulation = True
2026-06-07T13:07:40.656527 -
0%| | 0/6 [00:00
2026-06-07T13:12:56.228828 - [INFO] 🔍 Has 'model' attr? False
2026-06-07T13:12:56.228957 - [INFO] 🔍 Has 'loras' attr? False
2026-06-07T13:12:56.229086 - [INFO] 🔧 Wrapping NunchakuQwenImageTransformer2DModel with ComfyQwenImageWrapper
2026-06-07T13:12:56.229430 - [INFO] 📦 Creating ComfyQwenImageWrapper with cpu_offload='enable', apply_awq_mod=True (V3: always enabled)
2026-06-07T13:12:56.230625 - [INFO] 🔧 CPU offload setting: 'enable' (VRAM margin: 4.0GB)
2026-06-07T13:12:56.237056 - [INFO] Total LoRAs in stack: 2
2026-06-07T13:12:56.259599 - [INFO] Using sage attention mode: auto
2026-06-07T13:12:56.328643 - [INFO] Requested to load NunchakuQwenImage
2026-06-07T13:12:56.403388 - [INFO] Patching torch settings: torch.backends.cuda.matmul.allow_fp16_accumulation = True
2026-06-07T13:12:56.430051 -
0%| | 0/6 [00:000 else latent_image
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\AI\ComfyUI\nodes.py", line 1586, in sample
return common_ksampler(model, seed, steps, cfg, sampler_name, scheduler, positive, negative, latent_image, denoise=denoise)
File "D:\AI\ComfyUI\nodes.py", line 1550, 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:\AI\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:\AI\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:\AI\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:\AI\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:\AI\ComfyUI\comfy\patcher_extension.py", line 113, in execute
return self.original(*args, **kwargs)
~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "D:\AI\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:\AI\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:\AI\ComfyUI\comfy\patcher_extension.py", line 113, in execute
return self.original(*args, **kwargs)
~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "D:\AI\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:\AI\python-3.13.13\comfyui\Lib\site-packages\torch\utils\_contextlib.py", line 120, in decorate_context
return func(*args, **kwargs)
File "D:\AI\ComfyUI\comfy\k_diffusion\sampling.py", line 205, in sample_euler
denoised = model(x, sigma_hat * s_in, **extra_args)
File "D:\AI\ComfyUI\comfy\samplers.py", line 639, in __call__
out = self.inner_model(x, sigma, model_options=model_options, seed=seed)
File "D:\AI\ComfyUI\comfy\samplers.py", line 1202, in __call__
return self.outer_predict_noise(*args, **kwargs)
~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "D:\AI\ComfyUI\comfy\samplers.py", line 1209, in outer_predict_noise
).execute(x, timestep, model_options, seed)
~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\AI\ComfyUI\comfy\patcher_extension.py", line 113, in execute
return self.original(*args, **kwargs)
~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "D:\AI\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:\AI\ComfyUI\comfy\samplers.py", line 619, in sampling_function
out = calc_cond_batch(model, conds, x, timestep, model_options)
File "D:\AI\ComfyUI\comfy\samplers.py", line 210, in calc_cond_batch
return _calc_cond_batch_outer(model, conds, x_in, timestep, model_options)
File "D:\AI\ComfyUI\comfy\samplers.py", line 218, in _calc_cond_batch_outer
return executor.execute(model, conds, x_in, timestep, model_options)
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\AI\ComfyUI\comfy\patcher_extension.py", line 113, in execute
return self.original(*args, **kwargs)
~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "D:\AI\ComfyUI\comfy\samplers.py", line 334, in _calc_cond_batch
output = model.apply_model(input_x, timestep_, **c).chunk(batch_chunks)
~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\AI\ComfyUI\comfy\model_base.py", line 187, 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:\AI\ComfyUI\comfy\patcher_extension.py", line 113, in execute
return self.original(*args, **kwargs)
~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "D:\AI\ComfyUI\comfy\model_base.py", line 231, in _apply_model
model_output = self.diffusion_model(xc, t, context=context, control=control, transformer_options=transformer_options, **extra_conds)
File "D:\AI\python-3.13.13\comfyui\Lib\site-packages\torch\nn\modules\module.py", line 1775, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "D:\AI\python-3.13.13\comfyui\Lib\site-packages\torch\nn\modules\module.py", line 1786, in _call_impl
return forward_call(*args, **kwargs)
File "D:\AI\ComfyUI\custom_nodes\ComfyUI-QwenImageLoraLoader\wrappers\qwenimage.py", line 295, in forward
out = self._execute_model(x, timestep, context, guidance, control, transformer_options, **kwargs)
File "D:\AI\ComfyUI\custom_nodes\ComfyUI-QwenImageLoraLoader\wrappers\qwenimage.py", line 414, in _execute_model
return self.model(
~~~~~~~~~~^
x,
^^
...<6 lines>...
**final_kwargs_cleaned,
^^^^^^^^^^^^^^^^^^^^^^^
)
^
File "D:\AI\python-3.13.13\comfyui\Lib\site-packages\torch\nn\modules\module.py", line 1775, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "D:\AI\python-3.13.13\comfyui\Lib\site-packages\torch\nn\modules\module.py", line 1786, in _call_impl
return forward_call(*args, **kwargs)
File "D:\AI\ComfyUI\comfy\ldm\qwen_image\model.py", line 419, 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, timestep, context, attention_mask, ref_latents, additional_t_cond, transformer_options, **kwargs)
~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\AI\ComfyUI\comfy\patcher_extension.py", line 113, in execute
return self.original(*args, **kwargs)
~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "D:\AI\ComfyUI\custom_nodes\ComfyUI-nunchaku\models\qwenimage.py", line 813, in _forward
hidden_states = hidden_states[:, :num_embeds].view(
orig_shape[0], orig_shape[-2] // 2, orig_shape[-1] // 2, orig_shape[1], 2, 2
)
RuntimeError: shape '[1, 68, 120, 16, 2, 2]' is invalid for input of size 1044480

2026-06-07T13:13:22.078037 - [INFO] Prompt executed in 50.67 seconds

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

Workflow too large. Please manually upload the workflow from local file system.

## Additional Context
(Please add any additional context or steps to reproduce the error here)
```

### Other

_No response_