Blackwell (RTX 5060 Ti): intermittent Fault failed: 2 (cuMemSetAccess: device not ready) with DynamicVRAM/VBAR

Summary

I'm seeing intermittent crashes with DynamicVRAM/VBAR on a Blackwell GPU (RTX 5060 Ti 16 GB). The crash disappears completely when DynamicVRAM (or Smart Memory) is disabled, which seems to isolate the issue to AIMDO/VBAR rather than ComfyUI, the model, or VRAM exhaustion.

Environment

• ComfyUI: 0.34.2
• comfy-aimdo: 0.4.15
• Windows
• Python 3.13.13
• PyTorch 2.13.0+cu130
• CUDA 13
• GPU: NVIDIA GeForce RTX 5060 Ti 16 GB (Blackwell)

Launch command:

python main.py --listen --port 8188 --cuda-device 1 --use-ck-attention

Note: --cuda-device 1 is intentional because this is a dual-GPU system.

Error

The crash always happens during "Model Initializing..." before the first sampling step.

Relevant log:

CUDA API FAILED (600):

cuMemSetAccess(vaddr, size, &accessDesc, 1): device not ready


VRAM Allocation failed (non OOM)

RuntimeError: Fault failed: 2

The traceback points to:

comfy_aimdo/model_vbar.py

vbar_fault()

Reproduction

The issue is intermittent.

Typical pattern:

• several generations succeed
• a new generation starts
• "Model Initializing..." appears
• immediate crash with Fault failed: 2
• restarting the next generation usually works again, relaunching the same generation also works.

I can reproduce this repeatedly after generating multiple images with Krea2.

What I tested

Configuration | Result
-- | --
Default | ❌ Intermittent Fault failed: 2
--disable-smart-memory | ✅ No crashes
--disable-dynamic-vram | ✅ No crashes (slightly slower)

This strongly suggests the problem is related to DynamicVRAM/VBAR.

Additional notes

• This is not an OOM condition.
• The GPU still has around 15.9 GB free VRAM when the crash occurs.
• The error happens before sampling begins.
• The issue appears specific to DynamicVRAM because both --disable-smart-memory and --disable-dynamic-vram eliminate it.
• The model used was Krea2, but the traceback consistently points to VBAR memory fault handling rather than model execution itself.

the full ComfyUI error report and workflow.

Node threw an error during execution.

# ComfyUI Error Report
## Error Details
- **Node ID:** 117
- **Node Type:** KSampler
- **Exception Type:** RuntimeError
- **Exception Message:** RuntimeError: Fault failed: 2

## Stack Trace
```
File "D:\stablediffusion\AltComfy\ComfyUI\execution.py", line 545, 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:\stablediffusion\AltComfy\ComfyUI\execution.py", line 344, 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:\stablediffusion\AltComfy\ComfyUI\execution.py", line 318, in _async_map_node_over_list
await process_inputs(input_dict, i)

File "D:\stablediffusion\AltComfy\ComfyUI\execution.py", line 306, in process_inputs
result = f(**inputs)

File "D:\stablediffusion\AltComfy\ComfyUI\nodes.py", line 1623, in sample
return common_ksampler(model, seed, steps, cfg, sampler_name, scheduler, positive, negative, latent_image, denoise=denoise)

File "D:\stablediffusion\AltComfy\ComfyUI\nodes.py", line 1587, 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:\stablediffusion\AltComfy\ComfyUI\comfy\sample.py", line 82, 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:\stablediffusion\AltComfy\ComfyUI\comfy\samplers.py", line 1463, 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:\stablediffusion\AltComfy\ComfyUI\comfy\samplers.py", line 1353, in sample
return cfg_guider.sample(noise, latent_image, sampler, sigmas, denoise_mask, callback, disable_pbar, seed)
~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "D:\stablediffusion\AltComfy\ComfyUI\comfy\samplers.py", line 1335, in sample
output = executor.execute(noise, latent_image, sampler, sigmas, denoise_mask, callback, disable_pbar, seed, latent_shapes=latent_shapes)

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

File "D:\stablediffusion\AltComfy\ComfyUI\comfy\samplers.py", line 1262, 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:\stablediffusion\AltComfy\ComfyUI\comfy\samplers.py", line 1237, in inner_sample
samples = executor.execute(self, sigmas, extra_args, callback, noise, latent_image, denoise_mask, disable_pbar)

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

File "D:\stablediffusion\AltComfy\ComfyUI\comfy\samplers.py", line 1005, 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:\stablediffusion\AltComfy\.venv\Lib\site-packages\torch\utils\_contextlib.py", line 124, in decorate_context
return func(*args, **kwargs)

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

File "D:\stablediffusion\AltComfy\ComfyUI\comfy\samplers.py", line 640, in __call__
out = self.inner_model(x, sigma, model_options=model_options, seed=seed)

File "D:\stablediffusion\AltComfy\ComfyUI\comfy\samplers.py", line 1208, in __call__
return self.outer_predict_noise(*args, **kwargs)
~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^

File "D:\stablediffusion\AltComfy\ComfyUI\comfy\samplers.py", line 1215, in outer_predict_noise
).execute(x, timestep, model_options, seed)
~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

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

File "D:\stablediffusion\AltComfy\ComfyUI\comfy\samplers.py", line 1218, 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:\stablediffusion\AltComfy\ComfyUI\comfy\samplers.py", line 620, in sampling_function
out = calc_cond_batch(model, conds, x, timestep, model_options)

File "D:\stablediffusion\AltComfy\ComfyUI\comfy\samplers.py", line 211, in calc_cond_batch
return _calc_cond_batch_outer(model, conds, x_in, timestep, model_options)

File "D:\stablediffusion\AltComfy\ComfyUI\comfy\samplers.py", line 219, in _calc_cond_batch_outer
return executor.execute(model, conds, x_in, timestep, model_options)
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

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

File "D:\stablediffusion\AltComfy\ComfyUI\comfy\samplers.py", line 335, in _calc_cond_batch
output = model.apply_model(input_x, timestep_, **c).chunk(batch_chunks)
~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^

File "D:\stablediffusion\AltComfy\ComfyUI\comfy\model_base.py", line 205, 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:\stablediffusion\AltComfy\ComfyUI\comfy\patcher_extension.py", line 113, in execute
return self.original(*args, **kwargs)
~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^

File "D:\stablediffusion\AltComfy\ComfyUI\comfy\model_base.py", line 249, in _apply_model
model_output = self.diffusion_model(xc, t, context=context, control=control, transformer_options=transformer_options, **extra_conds)

File "D:\stablediffusion\AltComfy\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1778, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^

File "D:\stablediffusion\AltComfy\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1789, in _call_impl
return forward_call(*args, **kwargs)

File "D:\stablediffusion\AltComfy\ComfyUI\comfy\ldm\krea2\model.py", line 277, 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, ref_latents, transformer_options, **kwargs)
~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

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

File "D:\stablediffusion\AltComfy\ComfyUI\comfy\ldm\krea2\model.py", line 370, in _forward
combined = block(combined, tvec, freqs, None, timestep_zero_index=timestep_zero_index, transformer_options=transformer_options)

File "D:\stablediffusion\AltComfy\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1778, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^

File "D:\stablediffusion\AltComfy\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1789, in _call_impl
return forward_call(*args, **kwargs)

File "D:\stablediffusion\AltComfy\ComfyUI\comfy\ldm\krea2\model.py", line 214, in forward
x = x + pregate * self.attn((1 + prescale) * self.prenorm(x) + preshift, freqs, mask, transformer_options=transformer_options)
~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "D:\stablediffusion\AltComfy\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1778, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^

File "D:\stablediffusion\AltComfy\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1789, in _call_impl
return forward_call(*args, **kwargs)

File "D:\stablediffusion\AltComfy\ComfyUI\comfy\ldm\krea2\model.py", line 79, in forward
q, k, v, gate = self.wq(x), self.wk(x), self.wv(x), self.gate(x)
~~~~~~~^^^

File "D:\stablediffusion\AltComfy\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1778, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^

File "D:\stablediffusion\AltComfy\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1789, in _call_impl
return forward_call(*args, **kwargs)

File "D:\stablediffusion\AltComfy\ComfyUI\comfy\ops.py", line 1414, in forward
output = self.forward_comfy_cast_weights(
input,
...<2 lines>...
weight_only_quant=weight_only_quant,
)

File "D:\stablediffusion\AltComfy\ComfyUI\comfy\ops.py", line 1347, in forward_comfy_cast_weights
with CastBiasWeightContext(
~~~~~~~~~~~~~~~~~~~~~^
self,
^^^^^
...<6 lines>...
want_requant=True,
^^^^^^^^^^^^^^^^^^
) as (weight, bias):
^

File "D:\stablediffusion\AltComfy\ComfyUI\comfy\ops.py", line 466, in __init__
self.state = (None, None) if self.slf is None else cast_bias_weight(*args, **kwargs)
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^

File "D:\stablediffusion\AltComfy\ComfyUI\comfy\ops.py", line 377, in cast_bias_weight
offload_stream = cast_modules_with_vbar([s], dtype, device, bias_dtype, non_blocking)

File "D:\stablediffusion\AltComfy\ComfyUI\comfy\ops.py", line 166, in cast_modules_with_vbar
signature = comfy_aimdo.model_vbar.vbar_fault(s._v)

File "D:\stablediffusion\AltComfy\.venv\Lib\site-packages\comfy_aimdo\model_vbar.py", line 137, in vbar_fault
return vbar.fault(offset, size)
~~~~~~~~~~^^^^^^^^^^^^^^

File "D:\stablediffusion\AltComfy\.venv\Lib\site-packages\comfy_aimdo\model_vbar.py", line 94, in fault
raise RuntimeError(f"Fault failed: {res}")

```
## System Information
- **ComfyUI Version:** 0.34.2
- **Arguments:** main.py --listen --port 8188 --cuda-device 1 --use-ck-attention --disable-dynamic-vram
- **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:** 2.13.0+cu130
## Devices

- **Name:** cuda:0 NVIDIA GeForce RTX 5060 Ti : cudaMallocAsync
- **Type:** cuda
- **VRAM Total:** 17074421760
- **VRAM Free:** 15911092224
- **Torch VRAM Total:** 0
- **Torch VRAM Free:** 0

## Logs
```
2026-08-31T13:06:26.089544 - [INFO] setup plugin alembic.autogenerate.schemas
2026-08-31T13:06:26.089742 - [INFO] setup plugin alembic.autogenerate.tables
2026-08-31T13:06:26.090467 - [INFO] setup plugin alembic.autogenerate.types
2026-08-31T13:06:26.090811 - [INFO] setup plugin alembic.autogenerate.constraints
2026-08-31T13:06:26.091148 - [INFO] setup plugin alembic.autogenerate.defaults
2026-08-31T13:06:26.091496 - [INFO] setup plugin alembic.autogenerate.comments
2026-08-31T13:06:26.278059 - [INFO] Set cuda device to: 1
2026-08-31T13:06:26.950086 - [START] Security scan2026-08-31T13:06:26.950098 -
2026-08-31T13:06:28.534352 - [DONE] Security scan2026-08-31T13:06:28.534367 -
2026-08-31T13:06:28.705198 - ## ComfyUI-Manager: installing dependencies done.2026-08-31T13:06:28.705337 -
2026-08-31T13:06:28.705510 - ** ComfyUI startup time:2026-08-31T13:06:28.705837 - 2026-08-31T13:06:28.705953 - 2026-08-31 13:06:28.7052026-08-31T13:06:28.706071 -
2026-08-31T13:06:28.706199 - ** Platform:2026-08-31T13:06:28.706313 - 2026-08-31T13:06:28.706698 - Windows2026-08-31T13:06:28.706813 -
2026-08-31T13:06:28.706935 - ** Python version:2026-08-31T13:06:28.707204 - 2026-08-31T13:06:28.707319 - 3.13.13 (tags/v3.13.13:01104ce, Apr 7 2026, 19:25:48) [MSC v.1944 64 bit (AMD64)]2026-08-31T13:06:28.707432 -
2026-08-31T13:06:28.707545 - ** Python executable:2026-08-31T13:06:28.707810 - 2026-08-31T13:06:28.707907 - D:\stablediffusion\AltComfy\.venv\Scripts\python.exe2026-08-31T13:06:28.708020 -
2026-08-31T13:06:28.708140 - ** ComfyUI Path:2026-08-31T13:06:28.708441 - 2026-08-31T13:06:28.708539 - D:\stablediffusion\AltComfy\ComfyUI2026-08-31T13:06:28.708625 -
2026-08-31T13:06:28.708761 - ** ComfyUI Base Folder Path:2026-08-31T13:06:28.709163 - 2026-08-31T13:06:28.709274 - D:\stablediffusion\AltComfy\ComfyUI2026-08-31T13:06:28.709399 -
2026-08-31T13:06:28.709524 - ** User directory:2026-08-31T13:06:28.709880 - 2026-08-31T13:06:28.709990 - D:\stablediffusion\AltComfy\ComfyUI\user2026-08-31T13:06:28.710099 -
2026-08-31T13:06:28.710409 - ** ComfyUI-Manager config path:2026-08-31T13:06:28.710511 - 2026-08-31T13:06:28.710609 - D:\stablediffusion\AltComfy\ComfyUI\user\__manager\config.ini2026-08-31T13:06:28.710731 -
2026-08-31T13:06:28.710841 - ** Log path:2026-08-31T13:06:28.711136 - 2026-08-31T13:06:28.711238 - D:\stablediffusion\AltComfy\ComfyUI\user\comfyui.log2026-08-31T13:06:28.711340 -
2026-08-31T13:06:30.198008 - [INFO]
Prestartup times for custom nodes:
2026-08-31T13:06:30.198253 - [INFO] 0.0 seconds: D:\stablediffusion\AltComfy\ComfyUI\custom_nodes\ComfyUI-Easy-Use
2026-08-31T13:06:30.199062 - [INFO] 0.0 seconds: D:\stablediffusion\AltComfy\ComfyUI\custom_nodes\rgthree-comfy
2026-08-31T13:06:30.199408 - [INFO] 3.9 seconds: D:\stablediffusion\AltComfy\ComfyUI\custom_nodes\ComfyUI-Manager
2026-08-31T13:06:30.199783 - [INFO]
2026-08-31T13:06:34.985769 - [INFO] Found comfy_kitchen backend eager: {'available': True, 'disabled': False, 'unavailable_reason': None, 'capabilities': ['adaln', 'apply_rope', 'apply_rope1', 'apply_rope1_', 'apply_rope_', 'apply_rope_split_half', 'apply_rope_split_half1', 'apply_rope_split_half1_', 'apply_rope_split_half_', 'convrot_w4a4_linear', 'dequantize_convrot_w4a4_weight', 'dequantize_int8_convrot_weight', 'dequantize_int8_convrot_weight_dtype', 'dequantize_int8_embedding', 'dequantize_int8_simple', 'dequantize_int8_simple_dtype', 'dequantize_mxfp8', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'dequantize_w4a8_int8_weight', 'gemv_awq_w4a16', 'int8_linear', 'na3d', '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', 'quantize_w4a8_int8_weight', 'rms_adaln', 'rms_rope', 'rms_rope1', 'rms_rope1_', 'rms_rope_', 'rms_rope_split_half', 'rms_rope_split_half1', 'rms_rope_split_half1_', 'rms_rope_split_half_', 'rotate_int8_convrot_weight', 'scaled_mm_mxfp8', 'scaled_mm_nvfp4', 'scaled_mm_svdquant_w4a4', 'stochastic_rounding_fp8', 'w4a8_int8_linear']}
2026-08-31T13:06:34.986165 - [INFO] Found comfy_kitchen backend cuda: {'available': True, 'disabled': False, 'unavailable_reason': None, 'capabilities': ['adaln', 'apply_rope', 'apply_rope1', 'apply_rope1_', 'apply_rope_', 'apply_rope_split_half', 'apply_rope_split_half1', 'apply_rope_split_half1_', 'apply_rope_split_half_', 'convrot_w4a4_linear', 'dequantize_convrot_w4a4_weight', 'dequantize_int8_convrot_weight', 'dequantize_int8_convrot_weight_dtype', 'dequantize_int8_simple', 'dequantize_int8_simple_dtype', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'dequantize_w4a8_int8_weight', 'gemv_awq_w4a16', 'int8_linear', 'na3d', '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', 'quantize_w4a8_int8_weight', 'rms_adaln', 'rms_rope', 'rms_rope1', 'rms_rope1_', 'rms_rope_', 'rms_rope_split_half', 'rms_rope_split_half1', 'rms_rope_split_half1_', 'rms_rope_split_half_', 'rotate_int8_convrot_weight', 'scaled_mm_nvfp4', 'scaled_mm_svdquant_w4a4', 'stochastic_rounding_fp8', 'w4a8_int8_linear']}
2026-08-31T13:06:34.987458 - [INFO] Found comfy_kitchen backend triton: {'available': True, 'disabled': True, 'unavailable_reason': None, 'capabilities': ['adaln', 'apply_rope', 'apply_rope1', 'apply_rope1_', 'apply_rope_', 'apply_rope_split_half', 'apply_rope_split_half1', 'apply_rope_split_half1_', 'apply_rope_split_half_', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'int8_linear', 'na3d', 'quantize_and_rotate_rowwise', 'quantize_int8_rowwise', 'quantize_mxfp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8', 'rms_adaln', 'rms_rope', 'rms_rope1', 'rms_rope1_', 'rms_rope_', 'rms_rope_split_half', 'rms_rope_split_half1', 'rms_rope_split_half1_', 'rms_rope_split_half_', 'w4a8_int8_linear']}
2026-08-31T13:06:34.988033 - [INFO] Found comfy_kitchen backend hip: {'available': False, 'disabled': False, 'unavailable_reason': 'PyTorch ROCm/HIP runtime not available', 'capabilities': []}
2026-08-31T13:06:35.000594 - [INFO] Checkpoint files will always be loaded safely.
2026-08-31T13:06:35.150097 - [INFO] Total VRAM 16283 MB, total RAM 65456 MB
2026-08-31T13:06:35.150311 - [INFO] pytorch version: 2.13.0+cu130
2026-08-31T13:06:35.151986 - [INFO] Set vram state to: NORMAL_VRAM
2026-08-31T13:06:35.152241 - [INFO] Device: cuda:0 NVIDIA GeForce RTX 5060 Ti : cudaMallocAsync
2026-08-31T13:06:35.168603 - [INFO] Using async weight offloading with 2 streams
2026-08-31T13:06:35.171016 - [INFO] Enabled pinned memory 26182.0
2026-08-31T13:06:35.260729 - [INFO] Using pytorch attention
2026-08-31T13:06:35.261008 - [INFO] Using Comfy Kitchen attention
2026-08-31T13:06:42.196017 - [INFO] aimdo: src-win/cuda-detour.c:38:INFO:aimdo_setup_hooks: installing 6 hooks
2026-08-31T13:06:42.238808 - [INFO] aimdo: src/control.c:262:INFO:comfy-aimdo NVML pressure enabled
2026-08-31T13:06:42.242416 - [INFO] aimdo: src-win/shmem-detect.c:80:INFO:comfy-aimdo WDDM adapter match: NVIDIA GeForce RTX 5060 Ti runtime_luid=00000000:a0d03107 dxgi_luid=00000000:a0d03107
2026-08-31T13:06:42.242771 - [INFO] aimdo: src/control.c:277:INFO:comfy-aimdo inited for GPU: NVIDIA GeForce RTX 5060 Ti (VRAM: 16283 MB)
2026-08-31T13:06:42.243270 - [INFO] DynamicVRAM support detected and enabled
2026-08-31T13:06:42.243634 - [INFO] Python version: 3.13.13 (tags/v3.13.13:01104ce, Apr 7 2026, 19:25:48) [MSC v.1944 64 bit (AMD64)]
2026-08-31T13:06:42.244053 - [INFO] ComfyUI version: 0.34.2
2026-08-31T13:06:42.245913 - [INFO] comfy-aimdo version: 0.4.15
2026-08-31T13:06:42.248402 - [INFO] comfy-kitchen version: 0.2.31
2026-08-31T13:06:42.257415 - [INFO] comfyui-frontend-package version: 1.49.6
2026-08-31T13:06:42.257686 - [INFO] comfyui-workflow-templates version: 0.11.50
2026-08-31T13:06:42.258096 - [INFO] comfyui-embedded-docs version: 0.5.10
2026-08-31T13:06:42.258298 - [INFO] comfy-kitchen version: 0.2.31
2026-08-31T13:06:42.258685 - [INFO] comfy-aimdo version: 0.4.15
2026-08-31T13:06:42.262009 - [INFO] [Prompt Server] web root: D:\stablediffusion\AltComfy\.venv\Lib\site-packages\comfyui_frontend_package\static
2026-08-31T13:06:42.263171 - [INFO] Asset seeder disabled
2026-08-31T13:06:43.283964 - [INFO] No OpenGL_accelerate module loaded: No module named 'OpenGL_accelerate'
2026-08-31T13:06:50.343203 - [ComfyUI-Easy-Use] server: v1.3.7 Loaded2026-08-31T13:06:50.343392 -
2026-08-31T13:06:50.343571 - [ComfyUI-Easy-Use] web root: D:\stablediffusion\AltComfy\ComfyUI\custom_nodes\ComfyUI-Easy-Use\web_version/v2 Loaded2026-08-31T13:06:50.343985 -
2026-08-31T13:06:50.393491 - [INFO] ### Loading: ComfyUI-Impact-Pack (V8.28.3)
2026-08-31T13:06:50.666670 - [INFO] [Impact Pack] Wildcard total size (0.00 MB) is within cache limit (50.00 MB). Using full cache mode.
2026-08-31T13:06:50.668745 - [INFO] [Impact Pack] Wildcards loading done.
2026-08-31T13:06:51.379188 - D:\stablediffusion\AltComfy\.venv\Lib\site-packages\transparent_background\gui.py:24: UserWarning: Failed to import flet. Ignore this message when you do not need GUI mode.
warnings.warn('Failed to import flet. Ignore this message when you do not need GUI mode.')
2026-08-31T13:06:51.472524 - [INFO] ### Loading: ComfyUI-Manager (V3.41)
2026-08-31T13:06:51.473370 - [INFO] [ComfyUI-Manager] network_mode: public
2026-08-31T13:06:51.473654 - [INFO] [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'.
2026-08-31T13:06:51.726876 - [INFO] ### ComfyUI Revision: 5830 [169fcf35] *DETACHED | Released on '2026-08-27'
2026-08-31T13:06:52.325745 - [INFO] [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/alter-list.json
2026-08-31T13:06:52.400637 - [INFO] [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/model-list.json
2026-08-31T13:06:52.440968 - [INFO] [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/github-stats.json
2026-08-31T13:06:52.593925 - [INFO] [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/custom-node-list.json
2026-08-31T13:06:52.700281 - [INFO] [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/extension-node-map.json
2026-08-31T13:06:53.632356 -
2026-08-31T13:06:53.632529 - [rgthree-comfy] Loaded 48 extraordinary nodes. 🎉2026-08-31T13:06:53.633789 -
2026-08-31T13:06:53.633965 -
2026-08-31T13:06:53.636234 - [rgthree-comfy] ComfyUI's new Node 2.0 rendering may be incompatible with some rgthree-comfy nodes and features, breaking some rendering as well as losing the ability to access a node's properties (a vital part of many nodes). It also appears to run MUCH more slowly spiking CPU usage and causing jankiness and unresponsiveness, especially with large workflows. Personally I am not planning to use the new Nodes 2.0 and, unfortunately, am not able to invest the time to investigate and overhaul rgthree-comfy where needed. If you have issues when Nodes 2.0 is enabled, I'd urge you to switch it off as well and join me in hoping ComfyUI is not planning to deprecate the existing, stable canvas rendering all together.
2026-08-31T13:06:53.636600 -
2026-08-31T13:06:53.642127 - [INFO]
Import times for custom nodes:
2026-08-31T13:06:53.642468 - [INFO] 0.0 seconds: D:\stablediffusion\AltComfy\ComfyUI\custom_nodes\websocket_image_save.py
2026-08-31T13:06:53.642853 - [INFO] 0.0 seconds: D:\stablediffusion\AltComfy\ComfyUI\custom_nodes\comfyui-thundergnome
2026-08-31T13:06:53.643160 - [INFO] 0.0 seconds: D:\stablediffusion\AltComfy\ComfyUI\custom_nodes\comfyui-krea2-conditioning
2026-08-31T13:06:53.643539 - [INFO] 0.0 seconds: D:\stablediffusion\AltComfy\ComfyUI\custom_nodes\cg-use-everywhere
2026-08-31T13:06:53.643926 - [INFO] 0.0 seconds: D:\stablediffusion\AltComfy\ComfyUI\custom_nodes\ComfyUI-H3-FaceRefine
2026-08-31T13:06:53.644240 - [INFO] 0.0 seconds: D:\stablediffusion\AltComfy\ComfyUI\custom_nodes\ComfyUI-MiniMaxH3Mod
2026-08-31T13:06:53.644588 - [INFO] 0.0 seconds: D:\stablediffusion\AltComfy\ComfyUI\custom_nodes\ComfyUI-INT8-Fast
2026-08-31T13:06:53.644975 - [INFO] 0.0 seconds: D:\stablediffusion\AltComfy\ComfyUI\custom_nodes\ComfyUI-VFI
2026-08-31T13:06:53.645394 - [INFO] 0.0 seconds: D:\stablediffusion\AltComfy\ComfyUI\custom_nodes\comfyui-various
2026-08-31T13:06:53.645824 - [INFO] 0.0 seconds: D:\stablediffusion\AltComfy\ComfyUI\custom_nodes\ComfyUI-SolAttn_triton
2026-08-31T13:06:53.646217 - [INFO] 0.0 seconds: D:\stablediffusion\AltComfy\ComfyUI\custom_nodes\comfyui-frame-interpolation
2026-08-31T13:06:53.646643 - [INFO] 0.0 seconds: D:\stablediffusion\AltComfy\ComfyUI\custom_nodes\plaguekind-nodes
2026-08-31T13:06:53.646999 - [INFO] 0.1 seconds: D:\stablediffusion\AltComfy\ComfyUI\custom_nodes\rgthree-comfy
2026-08-31T13:06:53.647371 - [INFO] 0.1 seconds: D:\stablediffusion\AltComfy\ComfyUI\custom_nodes\comfyui-kjnodes
2026-08-31T13:06:53.647791 - [INFO] 0.1 seconds: D:\stablediffusion\AltComfy\ComfyUI\custom_nodes\ComfyUI-MiniMaxH3
2026-08-31T13:06:53.648213 - [INFO] 0.1 seconds: D:\stablediffusion\AltComfy\ComfyUI\custom_nodes\comfyui_nvidia_rtx_nodes
2026-08-31T13:06:53.648638 - [INFO] 0.1 seconds: D:\stablediffusion\AltComfy\ComfyUI\custom_nodes\comfyui-videohelpersuite
2026-08-31T13:06:53.648911 - [INFO] 0.3 seconds: D:\stablediffusion\AltComfy\ComfyUI\custom_nodes\ComfyUI-Impact-Pack
2026-08-31T13:06:53.649221 - [INFO] 0.4 seconds: D:\stablediffusion\AltComfy\ComfyUI\custom_nodes\comfyui-ollama
2026-08-31T13:06:53.649590 - [INFO] 0.6 seconds: D:\stablediffusion\AltComfy\ComfyUI\custom_nodes\ComfyUI-Spectrum-MiniMax-H3
2026-08-31T13:06:53.649855 - [INFO] 0.7 seconds: D:\stablediffusion\AltComfy\ComfyUI\custom_nodes\comfyui-inspyrenet-rembg
2026-08-31T13:06:53.650311 - [INFO] 0.8 seconds: D:\stablediffusion\AltComfy\ComfyUI\custom_nodes\ComfyUI-Manager
2026-08-31T13:06:53.650784 - [INFO] 5.5 seconds: D:\stablediffusion\AltComfy\ComfyUI\custom_nodes\ComfyUI-Easy-Use
2026-08-31T13:06:53.651123 - [INFO]
2026-08-31T13:06:53.654901 - [INFO] Context impl SQLiteImpl.
2026-08-31T13:06:53.655334 - [INFO] Will assume non-transactional DDL.
2026-08-31T13:06:53.994887 - [INFO] Using RAM pressure cache.
2026-08-31T13:06:53.995818 - [INFO] Starting server

2026-08-31T13:06:53.996566 - [INFO] To see the GUI go to: http://0.0.0.0:8188
2026-08-31T13:06:53.997075 - [INFO] To see the GUI go to: http://[::]:8188
2026-08-31T13:09:18.985479 - FETCH ComfyRegistry Data [DONE]2026-08-31T13:09:18.985633 -
2026-08-31T13:09:19.324018 - [INFO] [ComfyUI-Manager] default cache updated: https://api.comfy.org/nodes
2026-08-31T13:09:19.376472 - FETCH DATA from: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/custom-node-list.json2026-08-31T13:09:19.376621 - 2026-08-31T13:09:19.543803 - [DONE]2026-08-31T13:09:19.543989 -
2026-08-31T13:09:19.684552 - [INFO] [ComfyUI-Manager] All startup tasks have been completed.
2026-08-31T13:10:36.474343 - [INFO] got prompt
2026-08-31T13:10:36.521794 - [INFO] Using pytorch attention in VAE
2026-08-31T13:10:36.523886 - [INFO] Using pytorch attention in VAE
2026-08-31T13:10:36.698094 - [INFO] VAE load device: cuda:0, offload device: cpu, dtype: torch.bfloat16
2026-08-31T13:10:36.767960 - [INFO] Found quantization metadata version 1
2026-08-31T13:10:36.769019 - [INFO] Using MixedPrecisionOps for text encoder
2026-08-31T13:10:37.261712 - [INFO] CLIP/text encoder model load device: cuda:0, offload device: cpu, current: cpu, dtype: torch.float16
2026-08-31T13:10:37.269961 - [INFO] Requested to load Krea2TEModel_
2026-08-31T13:10:37.404129 - [INFO] Model Krea2TEModel_ prepared for dynamic VRAM loading. 4999MB Staged. 0 patches attached. Force pre-loaded 249 weights: 627 KB.
2026-08-31T13:10:37.604090 - [INFO] got prompt
2026-08-31T13:10:40.734300 - [INFO] Found quantization metadata version 1
2026-08-31T13:10:40.734700 - [INFO] Detected mixed precision quantization
2026-08-31T13:10:40.736250 - [INFO] Using mixed precision operations
2026-08-31T13:10:40.736515 - [INFO] Native ops: int8_tensorwise, convrot_w4a4, float8_e4m3fn, nvfp4, asym_w4a8_int8, float8_e5m2, mxfp8
2026-08-31T13:10:40.764138 - [INFO] model weight dtype torch.bfloat16, manual cast: torch.bfloat16
2026-08-31T13:10:40.769206 - [INFO] model_type FLUX
2026-08-31T13:10:41.085705 - [EasyUse] Load LORA: Krea2\rackhamz_v2.safetensors: model=0.250, clip=0.250, LBW=None, A=None, B=None2026-08-31T13:10:41.085877 -
2026-08-31T13:10:41.214984 - [EasyUse] Load LORA: Krea2\NC_Wyeth_TW_NC_Wyeth.safetensors: model=0.500, clip=0.500, LBW=None, A=None, B=None2026-08-31T13:10:41.215159 -
2026-08-31T13:10:41.302091 - [EasyUse] Load LORA: Krea2\Waterhouse_TW_JW_Waterhouse.safetensors: model=0.400, clip=0.400, LBW=None, A=None, B=None2026-08-31T13:10:41.302236 -
2026-08-31T13:10:41.375774 - [EasyUse] Load LORA: Krea2\rackhamz_v2.safetensors: model=0.450, clip=0.450, LBW=None, A=None, B=None2026-08-31T13:10:41.375976 -
2026-08-31T13:10:41.447635 - [EasyUse] Load LORA: Krea2\EstiloSinteticoChromaKrea.safetensors: model=1.000, clip=1.000, LBW=None, A=None, B=None2026-08-31T13:10:41.447821 -
2026-08-31T13:10:41.517838 - [EasyUse] Load LORA: Krea2\Muriel_2.safetensors: model=0.600, clip=0.600, LBW=None, A=None, B=None2026-08-31T13:10:41.517989 -
2026-08-31T13:10:41.616183 - [INFO] Requested to load Krea2TEModel_
2026-08-31T13:10:41.712977 - [INFO] Model Krea2TEModel_ prepared for dynamic VRAM loading. 4999MB Staged. 0 patches attached. Force pre-loaded 249 weights: 627 KB.
2026-08-31T13:10:43.393458 - [WARNING] Warning: TAESD previews enabled, but could not find models/vae_approx/lighttaew2_1
2026-08-31T13:10:43.394262 - [INFO] Requested to load Krea2
2026-08-31T13:10:43.494360 - [INFO] Model Krea2 prepared for dynamic VRAM loading. 12864MB Staged. 256 patches attached. Force pre-loaded 160 weights: 2824 KB.
2026-08-31T13:11:05.948144 -
100%|████████████████████████████████████████████████████████████████████████████████████| 8/8 [00:12<00:00, 2.00s/it]2026-08-31T13:11:05.948486 -
100%|████████████████████████████████████████████████████████████████████████████████████| 8/8 [00:12<00:00, 1.55s/it]2026-08-31T13:11:05.948632 -
2026-08-31T13:11:07.897178 - [INFO] Requested to load WanVAE
2026-08-31T13:11:07.900355 - [INFO] 0 models unloaded.
2026-08-31T13:11:07.920277 - [INFO] Model WanVAE prepared for dynamic VRAM loading. 241MB Staged. 0 patches attached. Force pre-loaded 60 weights: 61 KB.
2026-08-31T13:11:08.915325 - [INFO] Prompt executed in 32.42 seconds
2026-08-31T13:11:09.478387 - [INFO] Model Krea2TEModel_ prepared for dynamic VRAM loading. 4999MB Staged. 0 patches attached. Force pre-loaded 249 weights: 627 KB.
2026-08-31T13:11:11.264733 - [WARNING] Warning: TAESD previews enabled, but could not find models/vae_approx/lighttaew2_1
2026-08-31T13:11:11.267828 - [INFO] 0 models unloaded.
2026-08-31T13:11:11.355775 - [INFO] Model Krea2 prepared for dynamic VRAM loading. 12864MB Staged. 256 patches attached. Force pre-loaded 160 weights: 2824 KB.
2026-08-31T13:11:11.358671 -
0%| | 0/8 [00:00...
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:\stablediffusion\AltComfy\ComfyUI\comfy\patcher_extension.py", line 113, in execute
return self.original(*args, **kwargs)
~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "D:\stablediffusion\AltComfy\ComfyUI\comfy\model_base.py", line 249, in _apply_model
model_output = self.diffusion_model(xc, t, context=context, control=control, transformer_options=transformer_options, **extra_conds)
File "D:\stablediffusion\AltComfy\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1778, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "D:\stablediffusion\AltComfy\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1789, in _call_impl
return forward_call(*args, **kwargs)
File "D:\stablediffusion\AltComfy\ComfyUI\comfy\ldm\krea2\model.py", line 277, 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, ref_latents, transformer_options, **kwargs)
~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\stablediffusion\AltComfy\ComfyUI\comfy\patcher_extension.py", line 113, in execute
return self.original(*args, **kwargs)
~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "D:\stablediffusion\AltComfy\ComfyUI\comfy\ldm\krea2\model.py", line 370, in _forward
combined = block(combined, tvec, freqs, None, timestep_zero_index=timestep_zero_index, transformer_options=transformer_options)
File "D:\stablediffusion\AltComfy\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1778, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "D:\stablediffusion\AltComfy\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1789, in _call_impl
return forward_call(*args, **kwargs)
File "D:\stablediffusion\AltComfy\ComfyUI\comfy\ldm\krea2\model.py", line 214, in forward
x = x + pregate * self.attn((1 + prescale) * self.prenorm(x) + preshift, freqs, mask, transformer_options=transformer_options)
~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\stablediffusion\AltComfy\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1778, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "D:\stablediffusion\AltComfy\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1789, in _call_impl
return forward_call(*args, **kwargs)
File "D:\stablediffusion\AltComfy\ComfyUI\comfy\ldm\krea2\model.py", line 79, in forward
q, k, v, gate = self.wq(x), self.wk(x), self.wv(x), self.gate(x)
~~~~~~~^^^
File "D:\stablediffusion\AltComfy\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1778, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "D:\stablediffusion\AltComfy\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1789, in _call_impl
return forward_call(*args, **kwargs)
File "D:\stablediffusion\AltComfy\ComfyUI\comfy\ops.py", line 1414, in forward
output = self.forward_comfy_cast_weights(
input,
...<2 lines>...
weight_only_quant=weight_only_quant,
)
File "D:\stablediffusion\AltComfy\ComfyUI\comfy\ops.py", line 1347, in forward_comfy_cast_weights
with CastBiasWeightContext(
~~~~~~~~~~~~~~~~~~~~~^
self,
^^^^^
...<6 lines>...
want_requant=True,
^^^^^^^^^^^^^^^^^^
) as (weight, bias):
^
File "D:\stablediffusion\AltComfy\ComfyUI\comfy\ops.py", line 466, in __init__
self.state = (None, None) if self.slf is None else cast_bias_weight(*args, **kwargs)
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "D:\stablediffusion\AltComfy\ComfyUI\comfy\ops.py", line 377, in cast_bias_weight
offload_stream = cast_modules_with_vbar([s], dtype, device, bias_dtype, non_blocking)
File "D:\stablediffusion\AltComfy\ComfyUI\comfy\ops.py", line 166, in cast_modules_with_vbar
signature = comfy_aimdo.model_vbar.vbar_fault(s._v)
File "D:\stablediffusion\AltComfy\.venv\Lib\site-packages\comfy_aimdo\model_vbar.py", line 137, in vbar_fault
return vbar.fault(offset, size)
~~~~~~~~~~^^^^^^^^^^^^^^
File "D:\stablediffusion\AltComfy\.venv\Lib\site-packages\comfy_aimdo\model_vbar.py", line 94, in fault
raise RuntimeError(f"Fault failed: {res}")
RuntimeError: Fault failed: 2

2026-08-31T13:11:12.608172 - [INFO] Prompt executed in 3.37 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)

Start by reproducing the failure with the listed launch command and compare runs with DynamicVRAM or Smart Memory disabled. Read comfy_aimdo/model_vbar.py, especially vbar_fault(), alongside the cuMemSetAccess error and the Model Initializing phase. Done means identifying and correcting the intermittent Fault failed: 2 without disabling DynamicVRAM or Smart Memory.

Written by the indexing model from the issue text.