Memory management bug after latest update, 5090, QWEN

### 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

Something wrong with QWEN after latest update, i experienced it several times during the day - It's most likely a bug.

### Actual Behavior

### Steps to Reproduce

Using template workflow - text to image.

### Debug Logs

```powershell
Requested to load QwenImage
loaded partially 28799.448330841064 28794.956176757812 501
42%|██████████████████████████████████▍ | 13/31 [03:03<04:26, 14.81s/it][Crystools ERROR] Could not get GPU temperature. Turning off this feature. exception: access violation reading 0x0000018DB75FE8F8
[Crystools ERROR] Could not get GPU utilization. exception: access violation reading 0x0000018DB75FE7B8
[Crystools ERROR] Monitor of GPU is turning off.
42%|██████████████████████████████████▍ | 13/31 [04:02<05:36, 18.68s/it]
!!! Exception during processing !!! CUDA error: unknown error
Search for `cudaErrorUnknown' in https://docs.nvidia.com/cuda/cuda-runtime-api/group__CUDART__TYPES.html for more information.
CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1
Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions.

Traceback (most recent call last):
File "C:\comfyui\ComfyUI_final\ComfyUI\execution.py", line 496, 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, hidden_inputs=hidden_inputs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\execution.py", line 315, 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, hidden_inputs=hidden_inputs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\execution.py", line 289, in _async_map_node_over_list
await process_inputs(input_dict, i)
File "C:\comfyui\ComfyUI_final\ComfyUI\execution.py", line 277, in process_inputs
result = f(**inputs)
^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\nodes.py", line 1525, in sample
return common_ksampler(model, seed, steps, cfg, sampler_name, scheduler, positive, negative, latent_image, denoise=denoise)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\nodes.py", line 1492, in common_ksampler
samples = comfy.sample.sample(model, noise, steps, cfg, sampler_name, scheduler, positive, negative, latent_image,
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\comfy\sample.py", line 45, 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 "C:\comfyui\ComfyUI_final\ComfyUI\comfy\samplers.py", line 1161, 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 "C:\comfyui\ComfyUI_final\ComfyUI\comfy\samplers.py", line 1051, in sample
return cfg_guider.sample(noise, latent_image, sampler, sigmas, denoise_mask, callback, disable_pbar, seed)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\comfy\samplers.py", line 1036, in sample
output = executor.execute(noise, latent_image, sampler, sigmas, denoise_mask, callback, disable_pbar, seed)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\comfy\patcher_extension.py", line 112, in execute
return self.original(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\comfy\samplers.py", line 1004, in outer_sample
output = self.inner_sample(noise, latent_image, device, sampler, sigmas, denoise_mask, callback, disable_pbar, seed)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\comfy\samplers.py", line 987, in inner_sample
samples = executor.execute(self, sigmas, extra_args, callback, noise, latent_image, denoise_mask, disable_pbar)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\comfy\patcher_extension.py", line 112, in execute
return self.original(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\comfy\samplers.py", line 759, in sample
samples = self.sampler_function(model_k, noise, sigmas, extra_args=extra_args, callback=k_callback, disable=disable_pbar, **self.extra_options)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\venv\Lib\site-packages\torch\utils\_contextlib.py", line 120, in decorate_context
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui/ComfyUI_final/ComfyUI\comfy\k_diffusion\sampling.py", line 199, in sample_euler
denoised = model(x, sigma_hat * s_in, **extra_args)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\comfy\samplers.py", line 408, in __call__
out = self.inner_model(x, sigma, model_options=model_options, seed=seed)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\comfy\samplers.py", line 960, in __call__
return self.outer_predict_noise(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\comfy\samplers.py", line 967, in outer_predict_noise
).execute(x, timestep, model_options, seed)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\comfy\patcher_extension.py", line 112, in execute
return self.original(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\comfy\samplers.py", line 970, 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 "C:\comfyui\ComfyUI_final\ComfyUI\comfy\samplers.py", line 388, in sampling_function
out = calc_cond_batch(model, conds, x, timestep, model_options)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\comfy\samplers.py", line 206, in calc_cond_batch
return _calc_cond_batch_outer(model, conds, x_in, timestep, model_options)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\comfy\samplers.py", line 214, in _calc_cond_batch_outer
return executor.execute(model, conds, x_in, timestep, model_options)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\comfy\patcher_extension.py", line 112, in execute
return self.original(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\comfy\samplers.py", line 333, in _calc_cond_batch
output = model.apply_model(input_x, timestep_, **c).chunk(batch_chunks)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\comfy\model_base.py", line 160, in apply_model
return comfy.patcher_extension.WrapperExecutor.new_class_executor(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\comfy\patcher_extension.py", line 112, in execute
return self.original(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\comfy\model_base.py", line 199, in _apply_model
model_output = self.diffusion_model(xc, t, context=context, control=control, transformer_options=transformer_options, **extra_conds).float()
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\venv\Lib\site-packages\torch\nn\modules\module.py", line 1775, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\venv\Lib\site-packages\torch\nn\modules\module.py", line 1786, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\comfy\ldm\qwen_image\model.py", line 363, in forward
return comfy.patcher_extension.WrapperExecutor.new_class_executor(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\comfy\patcher_extension.py", line 112, in execute
return self.original(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\comfy\ldm\qwen_image\model.py", line 446, in _forward
encoder_hidden_states, hidden_states = block(
^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\venv\Lib\site-packages\torch\nn\modules\module.py", line 1775, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\venv\Lib\site-packages\torch\nn\modules\module.py", line 1786, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\comfy\ldm\qwen_image\model.py", line 242, in forward
img_attn_output, txt_attn_output = self.attn(
^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\venv\Lib\site-packages\torch\nn\modules\module.py", line 1775, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\venv\Lib\site-packages\torch\nn\modules\module.py", line 1786, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\comfy\ldm\qwen_image\model.py", line 140, in forward
img_key = self.to_k(hidden_states).unflatten(-1, (self.heads, -1))
^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\venv\Lib\site-packages\torch\nn\modules\module.py", line 1775, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\venv\Lib\site-packages\torch\nn\modules\module.py", line 1786, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\comfy\ops.py", line 113, in forward
return self.forward_comfy_cast_weights(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\comfy\ops.py", line 108, in forward_comfy_cast_weights
weight, bias = cast_bias_weight(self, input)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\comfy\ops.py", line 88, in cast_bias_weight
weight = comfy.model_management.cast_to(s.weight, dtype, device, non_blocking=non_blocking, copy=has_function, stream=offload_stream)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\comfyui\ComfyUI_final\ComfyUI\comfy\model_management.py", line 1072, in cast_to
r.copy_(weight, non_blocking=non_blocking)
torch.AcceleratorError: CUDA error: unknown error
Search for `cudaErrorUnknown' in https://docs.nvidia.com/cuda/cuda-runtime-api/group__CUDART__TYPES.html for more information.
CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1
Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions.
```

### Other

_No response_

Start with the traceback entry points in comfy/ldm/qwen_image/model.py, comfy/samplers.py, and execution.py, then reproduce the template text-to-image workflow with the reported Qwen model. Check whether the CUDA error and memory-management symptoms persist after the latest update; done means identifying a reproducible regression and confirming its resolution with the same workflow.

Written by the indexing model from the issue text.