小白

# ComfyUI Error Report
## Error Details
- **Node ID:** 8
- **Node Type:** VAEDecode
- **Exception Type:** RuntimeError
- **Exception Message:** MPS backend out of memory (MPS allocated: 7.27 GiB, other allocations: 80.14 MiB, max allowed: 9.07 GiB). Tried to allocate 2.06 GiB on private pool. Use PYTORCH_MPS_HIGH_WATERMARK_RATIO=0.0 to disable upper limit for memory allocations (may cause system failure).

## Stack Trace
```
File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/execution.py", line 535, 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 "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/execution.py", line 335, 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 "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/execution.py", line 309, in _async_map_node_over_list
await process_inputs(input_dict, i)

File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/execution.py", line 297, in process_inputs
result = f(**inputs)
^^^^^^^^^^^

File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/nodes.py", line 316, in decode
images = vae.decode(latent)
^^^^^^^^^^^^^^^^^^

File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/comfy/sd.py", line 1013, in decode
model_management.raise_non_oom(e)

File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/comfy/model_management.py", line 286, in raise_non_oom
raise e

File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/comfy/sd.py", line 1006, in decode
out = self.first_stage_model.decode(samples, **vae_options).to(device=self.output_device, dtype=self.vae_output_dtype(), copy=True)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/comfy/ldm/models/autoencoder.py", line 254, in decode
dec = self.decoder(dec, **decoder_kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1776, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1787, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/comfy/ldm/modules/diffusionmodules/model.py", line 808, in forward
h1 = self.up[i_level].block[i_block](h1, temb, conv_carry_in, conv_carry_out, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1776, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1787, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/comfy/ldm/modules/diffusionmodules/model.py", line 214, in forward
h = self.norm1(h)
^^^^^^^^^^^^^

File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1776, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1787, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/comfy/ops.py", line 563, in forward
return super().forward(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/nn/modules/normalization.py", line 327, in forward
return F.group_norm(input, self.num_groups, self.weight, self.bias, self.eps)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/nn/functional.py", line 2990, in group_norm
return torch.group_norm(
^^^^^^^^^^^^^^^^^

File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/_refs/__init__.py", line 3356, in native_group_norm
out = (input_reshaped - mean) * rstd
~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~

```
## System Information
- **ComfyUI Version:** 0.21.1
- **Arguments:** /Applications/ComfyUI.app/Contents/Resources/ComfyUI/main.py --user-directory /Volumes/一个张张盘/confyui/user --input-directory /Volumes/一个张张盘/confyui/input --output-directory /Volumes/一个张张盘/confyui/output --front-end-root /Applications/ComfyUI.app/Contents/Resources/ComfyUI/web_custom_versions/desktop_app --base-directory /Volumes/一个张张盘/confyui --database-url sqlite:////Volumes/一个张张盘/confyui/user/comfyui.db --extra-model-paths-config /Users/lanyu/Library/Application Support/ComfyUI/extra_models_config.yaml --log-stdout --listen 127.0.0.1 --port 8000 --enable-manager
- **OS:** darwin
- **Python Version:** 3.12.11 (main, Aug 18 2025, 19:02:39) [Clang 20.1.4 ]
- **Embedded Python:** false
- **PyTorch Version:** 2.10.0
## Devices

- **Name:** mps
- **Type:** mps
- **VRAM Total:** 8589934592
- **VRAM Free:** 2206728192
- **Torch VRAM Total:** 8589934592
- **Torch VRAM Free:** 2206728192

## Logs
```
2026-05-17T20:05:52.958412 - setup plugin alembic.autogenerate.schemas
2026-05-17T20:05:52.958503 - setup plugin alembic.autogenerate.tables
2026-05-17T20:05:52.958546 - setup plugin alembic.autogenerate.types
2026-05-17T20:05:52.958577 - setup plugin alembic.autogenerate.constraints
2026-05-17T20:05:52.958604 - setup plugin alembic.autogenerate.defaults
2026-05-17T20:05:52.958631 - setup plugin alembic.autogenerate.comments
2026-05-17T20:05:53.660003 - Could not autodetect AIMDO implementation, assuming Nvidia
2026-05-17T20:05:53.660237 - comfy-aimdo unsupported operating system: Darwin
2026-05-17T20:05:53.660266 - NOTE: comfy-aimdo currently only supports Windows and Linux
2026-05-17T20:05:54.405895 - Adding extra search path custom_nodes /Volumes/一个张张盘/confyui/custom_nodes
2026-05-17T20:05:54.405955 - Adding extra search path download_model_base /Volumes/一个张张盘/confyui/models
2026-05-17T20:05:54.405982 - Adding extra search path custom_nodes /Applications/ComfyUI.app/Contents/Resources/ComfyUI/custom_nodes
2026-05-17T20:05:54.406008 - Setting output directory to: /Volumes/一个张张盘/confyui/output
2026-05-17T20:05:54.406036 - Setting input directory to: /Volumes/一个张张盘/confyui/input
2026-05-17T20:05:54.406051 - Setting user directory to: /Volumes/一个张张盘/confyui/user
2026-05-17T20:05:55.886083 - [START] Security scan2026-05-17T20:05:55.886100 -
2026-05-17T20:05:56.412052 - [ComfyUI-Manager] Using uv as Python module for pip operations.
2026-05-17T20:05:56.949005 - [DONE] Security scan2026-05-17T20:05:56.949038 -
2026-05-17T20:05:57.002549 - ** ComfyUI startup time:2026-05-17T20:05:57.002663 - 2026-05-17T20:05:57.002734 - 2026-05-17 20:05:56.9892026-05-17T20:05:57.002851 -
2026-05-17T20:05:57.002899 - ** Platform:2026-05-17T20:05:57.002921 - 2026-05-17T20:05:57.002943 - Darwin2026-05-17T20:05:57.002961 -
2026-05-17T20:05:57.002980 - ** Python version:2026-05-17T20:05:57.002997 - 2026-05-17T20:05:57.003014 - 3.12.11 (main, Aug 18 2025, 19:02:39) [Clang 20.1.4 ]2026-05-17T20:05:57.003031 -
2026-05-17T20:05:57.003049 - ** Python executable:2026-05-17T20:05:57.003066 - 2026-05-17T20:05:57.003084 - /Volumes/一个张张盘/confyui/.venv/bin/python2026-05-17T20:05:57.003100 -
2026-05-17T20:05:57.003147 - ** ComfyUI Path:2026-05-17T20:05:57.003163 - 2026-05-17T20:05:57.003179 - /Applications/ComfyUI.app/Contents/Resources/ComfyUI2026-05-17T20:05:57.003194 -
2026-05-17T20:05:57.003210 - ** ComfyUI Base Folder Path:2026-05-17T20:05:57.003230 - 2026-05-17T20:05:57.003246 - /Applications/ComfyUI.app/Contents/Resources/ComfyUI2026-05-17T20:05:57.003261 -
2026-05-17T20:05:57.003277 - ** User directory:2026-05-17T20:05:57.003293 - 2026-05-17T20:05:57.003316 - /Volumes/一个张张盘/confyui/user2026-05-17T20:05:57.003332 -
2026-05-17T20:05:57.003348 - ** ComfyUI-Manager config path:2026-05-17T20:05:57.003383 - 2026-05-17T20:05:57.003435 - /Volumes/一个张张盘/confyui/user/__manager/config.ini2026-05-17T20:05:57.003462 -
2026-05-17T20:05:57.003491 - ** Log path:2026-05-17T20:05:57.003510 - 2026-05-17T20:05:57.003529 - /Volumes/一个张张盘/confyui/user/comfyui.log2026-05-17T20:05:57.003548 -
2026-05-17T20:05:57.153647 - [ComfyUI-Manager] Skipped fixing the 'comfyui-frontend-package' dependency because the ComfyUI is outdated.
2026-05-17T20:05:57.154406 - [PRE] ComfyUI-Manager
2026-05-17T20:06:07.848305 - Found comfy_kitchen backend triton: {'available': False, 'disabled': True, 'unavailable_reason': "ImportError: No module named 'triton'", 'capabilities': []}
2026-05-17T20:06:07.848431 - Found comfy_kitchen backend cuda: {'available': False, 'disabled': True, 'unavailable_reason': 'Extension file not found: /Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/comfy_kitchen/backends/cuda/_C.abi3.so', 'capabilities': []}
2026-05-17T20:06:07.848491 - Found comfy_kitchen backend eager: {'available': True, 'disabled': False, 'unavailable_reason': None, 'capabilities': ['apply_rope', 'apply_rope1', 'dequantize_mxfp8', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'quantize_mxfp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8', 'scaled_mm_mxfp8', 'scaled_mm_nvfp4']}
2026-05-17T20:06:07.889851 - Checkpoint files will always be loaded safely.
2026-05-17T20:06:07.993951 - Total VRAM 8192 MB, total RAM 8192 MB
2026-05-17T20:06:07.994078 - pytorch version: 2.10.0
2026-05-17T20:06:07.997711 - Mac Version (26, 3, 1)
2026-05-17T20:06:08.015433 - Set vram state to: SHARED
2026-05-17T20:06:08.015588 - Device: mps
2026-05-17T20:06:10.478187 - Using sub quadratic optimization for attention, if you have memory or speed issues try using: --use-split-cross-attention
2026-05-17T20:06:57.948049 - Python version: 3.12.11 (main, Aug 18 2025, 19:02:39) [Clang 20.1.4 ]
2026-05-17T20:06:57.964296 - ComfyUI version: 0.21.1
2026-05-17T20:06:57.992995 - comfy-aimdo version: 0.3.0
2026-05-17T20:06:57.994247 - comfy-kitchen version: 0.2.8
2026-05-17T20:06:58.048426 - [Prompt Server] web root: /Applications/ComfyUI.app/Contents/Resources/ComfyUI/web_custom_versions/desktop_app
2026-05-17T20:06:58.049454 - Asset seeder disabled
2026-05-17T20:06:58.049585 - [START] ComfyUI-Manager
2026-05-17T20:06:58.471228 - [ComfyUI-Manager] Using GitPython backend2026-05-17T20:06:58.471307 -
2026-05-17T20:06:58.572500 - [ComfyUI-Manager] network_mode: public
2026-05-17T20:06:58.790023 - [ComfyUI-Manager] The matrix sharing feature has been disabled because the `matrix-nio` dependency is not installed.
To use this feature, please run the following command:
/Volumes/一个张张盘/confyui/.venv/bin/python -m pip install matrix-nio

2026-05-17T20:07:05.305683 -
2026-05-17T20:07:05.305791 - [rgthree-comfy] Loaded 48 exciting nodes. 🎉2026-05-17T20:07:05.305818 -
2026-05-17T20:07:05.305840 -
2026-05-17T20:07:05.305877 - [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.
2026-05-17T20:07:05.305899 -
2026-05-17T20:07:15.359620 - [ComfyUI-Easy-Use] server: v1.3.6 Loaded2026-05-17T20:07:15.359666 -
2026-05-17T20:07:15.359715 - [ComfyUI-Easy-Use] web root: /Volumes/一个张张盘/confyui/custom_nodes/comfyui-easy-use/web_version/v2 Loaded2026-05-17T20:07:15.359737 -
2026-05-17T20:07:18.998626 -
Import times for custom nodes:
2026-05-17T20:07:19.041575 - 0.0 seconds: /Applications/ComfyUI.app/Contents/Resources/ComfyUI/custom_nodes/websocket_image_save.py
2026-05-17T20:07:19.041748 - 0.0 seconds: /Volumes/一个张张盘/confyui/custom_nodes/comfyui-mxtoolkit
2026-05-17T20:07:19.041809 - 0.1 seconds: /Volumes/一个张张盘/confyui/custom_nodes/comfyui_essentials
2026-05-17T20:07:19.041847 - 0.1 seconds: /Volumes/一个张张盘/confyui/custom_nodes/comfyui-frame-interpolation
2026-05-17T20:07:19.041887 - 0.2 seconds: /Volumes/一个张张盘/confyui/custom_nodes/rgthree-comfy
2026-05-17T20:07:19.041923 - 0.3 seconds: /Volumes/一个张张盘/confyui/custom_nodes/comfyui-kjnodes
2026-05-17T20:07:19.041953 - 0.4 seconds: /Volumes/一个张张盘/confyui/custom_nodes/whiterabbit
2026-05-17T20:07:19.041984 - 3.0 seconds: /Volumes/一个张张盘/confyui/custom_nodes/comfyui-videohelpersuite
2026-05-17T20:07:19.042015 - 9.7 seconds: /Volumes/一个张张盘/confyui/custom_nodes/comfyui-easy-use
2026-05-17T20:07:19.042046 -
2026-05-17T20:07:19.052974 - Context impl SQLiteImpl.
2026-05-17T20:07:19.053142 - Will assume non-transactional DDL.
2026-05-17T20:07:19.404287 - Starting server

2026-05-17T20:07:19.404621 - To see the GUI go to: http://127.0.0.1:8000
2026-05-17T20:07:20.852550 - [DEPRECATION WARNING] Detected import of deprecated legacy API: /extensions/core/widgetInputs.js. This is likely caused by a custom node extension using outdated APIs. Please update your extensions or contact the extension author for an updated version.
2026-05-17T20:07:22.360553 - [ComfyUI-Manager] The ComfyRegistry cache update is still in progress, so an outdated cache is being used.
2026-05-17T20:07:22.517195 - FETCH DATA from: /Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/comfyui_manager/custom-node-list.json2026-05-17T20:07:22.517229 - 2026-05-17T20:07:22.564260 - [DONE]2026-05-17T20:07:22.564309 -
2026-05-17T20:08:48.474627 - got prompt
2026-05-17T20:08:52.664846 - model weight dtype torch.float16, manual cast: None
2026-05-17T20:08:52.966050 - model_type EPS
2026-05-17T20:09:16.139153 - Using split attention in VAE
2026-05-17T20:09:16.348261 - Using split attention in VAE
2026-05-17T20:09:20.208755 - VAE load device: mps, offload device: cpu, dtype: torch.bfloat16
2026-05-17T20:09:25.418722 - Requested to load SD1ClipModel
2026-05-17T20:09:25.528426 - loaded completely; 235.84 MB loaded, full load: True
2026-05-17T20:09:25.533445 - CLIP/text encoder model load device: cpu, offload device: cpu, current: cpu, dtype: torch.float16
2026-05-17T20:09:25.534607 - loaded diffusion model directly to GPU
2026-05-17T20:09:25.536431 - Requested to load BaseModel
2026-05-17T20:09:25.600089 - 0 models unloaded.
2026-05-17T20:09:25.636281 - loaded completely; 1639.41 MB loaded, full load: True
2026-05-17T20:10:10.712492 -
65%|██████▌ | 13/20 [00:42<00:16, 2.30s/it]2026-05-17T20:10:12.093459 - FETCH ComfyRegistry Data [DONE]
2026-05-17T20:10:12.769860 - [ComfyUI-Manager] default cache updated: https://api.comfy.org/nodes
2026-05-17T20:10:13.089547 - FETCH DATA from: /Volumes/一个张张盘/confyui/user/__manager/cache/1514988643_custom-node-list.json2026-05-17T20:10:13.089610 - 2026-05-17T20:10:13.158395 - [DONE]2026-05-17T20:10:13.158469 -
2026-05-17T20:10:13.190681 - [ComfyUI-Manager] All startup tasks have been completed.
2026-05-17T20:10:28.300942 -
100%|██████████| 20/20 [00:59<00:00, 2.32s/it]2026-05-17T20:10:28.301592 -
100%|██████████| 20/20 [00:59<00:00, 2.99s/it]2026-05-17T20:10:28.301627 -
2026-05-17T20:10:28.627359 - Requested to load AutoencoderKL
2026-05-17T20:10:28.987873 - Unloaded partially: 837.79 MB freed, 801.63 MB remains loaded, 28.13 MB buffer reserved, lowvram patches: 0
2026-05-17T20:10:29.050442 - Unloaded partially: 421.91 MB freed, 379.71 MB remains loaded, 28.13 MB buffer reserved, lowvram patches: 0
2026-05-17T20:10:29.051073 - 0 models unloaded.
2026-05-17T20:10:29.557897 - loaded completely; 159.56 MB loaded, full load: True
2026-05-17T20:10:39.761260 - Prompt executed in 111.18 seconds
2026-05-17T20:11:08.944902 - got prompt
2026-05-17T20:11:10.541127 - 0 models unloaded.
2026-05-17T20:11:15.196659 - loaded completely; 1639.41 MB loaded, full load: True
2026-05-17T20:12:03.857531 -
100%|██████████| 20/20 [00:48<00:00, 2.15s/it]2026-05-17T20:12:03.858168 -
100%|██████████| 20/20 [00:48<00:00, 2.41s/it]2026-05-17T20:12:03.858199 -
2026-05-17T20:12:03.904995 - Requested to load AutoencoderKL
2026-05-17T20:12:04.270895 - Unloaded partially: 781.53 MB freed, 857.88 MB remains loaded, 28.13 MB buffer reserved, lowvram patches: 0
2026-05-17T20:12:04.424290 - Unloaded partially: 632.86 MB freed, 225.02 MB remains loaded, 56.25 MB buffer reserved, lowvram patches: 0
2026-05-17T20:12:04.424686 - 0 models unloaded.
2026-05-17T20:12:04.649600 - loaded completely; 159.56 MB loaded, full load: True
2026-05-17T20:12:07.516476 - Prompt executed in 58.46 seconds
2026-05-18T17:30:26.867745 - got prompt
2026-05-18T17:30:33.352132 - 0 models unloaded.
2026-05-18T17:30:48.600095 - loaded completely; 1639.41 MB loaded, full load: True
2026-05-18T17:31:36.072176 -
100%|██████████| 20/20 [00:46<00:00, 1.98s/it]2026-05-18T17:31:36.072566 -
100%|██████████| 20/20 [00:46<00:00, 2.34s/it]2026-05-18T17:31:36.072610 -
2026-05-18T17:31:36.177111 - Requested to load AutoencoderKL
2026-05-18T17:31:36.603458 - Unloaded partially: 653.33 MB freed, 986.09 MB remains loaded, 25.02 MB buffer reserved, lowvram patches: 0
2026-05-18T17:31:36.623993 - Unloaded partially: 50.04 MB freed, 936.05 MB remains loaded, 25.02 MB buffer reserved, lowvram patches: 0
2026-05-18T17:31:36.624262 - 0 models unloaded.
2026-05-18T17:31:36.756161 - loaded completely; 159.56 MB loaded, full load: True
2026-05-18T17:31:39.740980 - Prompt executed in 72.75 seconds
2026-05-18T17:32:09.462193 - got prompt
2026-05-18T17:32:09.638572 - Prompt executed in 0.04 seconds
2026-05-18T17:32:31.217169 - got prompt
2026-05-18T17:32:31.224182 - Prompt executed in 0.00 seconds
2026-05-18T17:33:00.139456 - got prompt
2026-05-18T17:33:00.280273 - Prompt executed in 0.00 seconds
2026-05-18T17:33:14.008218 - got prompt
2026-05-18T17:33:16.034751 - loaded completely; 1639.41 MB loaded, full load: True
2026-05-18T17:33:45.654627 -
46%|████▌ | 11/24 [00:29<00:32, 2.48s/it]2026-05-18T17:33:47.967084 - got prompt
2026-05-18T17:34:13.464177 -
100%|██████████| 24/24 [00:57<00:00, 1.97s/it]2026-05-18T17:34:13.464528 -
100%|██████████| 24/24 [00:57<00:00, 2.39s/it]2026-05-18T17:34:13.464563 -
2026-05-18T17:34:13.557330 - Requested to load AutoencoderKL
2026-05-18T17:34:13.983412 - Unloaded partially: 728.39 MB freed, 911.03 MB remains loaded, 25.02 MB buffer reserved, lowvram patches: 0
2026-05-18T17:34:14.012366 - Unloaded partially: 137.53 MB freed, 773.50 MB remains loaded, 28.13 MB buffer reserved, lowvram patches: 0
2026-05-18T17:34:14.012641 - 0 models unloaded.
2026-05-18T17:34:14.246165 - loaded completely; 159.56 MB loaded, full load: True
2026-05-18T17:34:17.398387 - Prompt executed in 63.26 seconds
2026-05-18T17:34:20.859162 - loaded completely; 1639.41 MB loaded, full load: True
2026-05-18T17:35:31.854245 -
100%|██████████| 24/24 [01:10<00:00, 2.24s/it]2026-05-18T17:35:31.854716 -
100%|██████████| 24/24 [01:10<00:00, 2.95s/it]2026-05-18T17:35:31.854748 -
2026-05-18T17:35:31.914254 - Requested to load AutoencoderKL
2026-05-18T17:35:32.327378 - Unloaded partially: 865.91 MB freed, 773.50 MB remains loaded, 28.13 MB buffer reserved, lowvram patches: 0
2026-05-18T17:35:32.671642 - Unloaded partially: 717.24 MB freed, 56.26 MB remains loaded, 56.25 MB buffer reserved, lowvram patches: 0
2026-05-18T17:35:32.688614 - 0 models unloaded.
2026-05-18T17:35:33.455140 - loaded completely; 159.56 MB loaded, full load: True
2026-05-18T17:35:36.411664 - Prompt executed in 77.57 seconds
2026-05-18T17:36:28.029048 - got prompt
2026-05-18T17:36:28.348382 - Prompt executed in 0.09 seconds
2026-05-18T17:38:35.988943 - got prompt
2026-05-18T17:38:46.163662 - model weight dtype torch.float16, manual cast: None
2026-05-18T17:38:46.523363 - model_type EPS
2026-05-18T17:39:54.264344 - Using split attention in VAE
2026-05-18T17:39:54.485878 - Using split attention in VAE
2026-05-18T17:39:58.435893 - VAE load device: mps, offload device: cpu, dtype: torch.bfloat16
2026-05-18T17:40:26.464288 - Requested to load SDXLClipModel
2026-05-18T17:40:26.720820 - loaded completely; 1560.80 MB loaded, full load: True
2026-05-18T17:40:26.726079 - CLIP/text encoder model load device: cpu, offload device: cpu, current: cpu, dtype: torch.float16
2026-05-18T17:40:26.726238 - loaded diffusion model directly to GPU
2026-05-18T17:40:26.726598 - Requested to load SDXL
2026-05-18T17:40:27.054767 - loaded completely; 4897.05 MB loaded, full load: True
2026-05-18T18:02:54.642059 -
100%|██████████| 24/24 [22:22<00:00, 47.46s/it]2026-05-18T18:02:54.645659 -
100%|██████████| 24/24 [22:23<00:00, 55.96s/it]2026-05-18T18:02:54.645695 -
2026-05-18T18:02:54.773312 - Requested to load AutoencoderKL
2026-05-18T18:03:14.632461 - Unloaded partially: 559.63 MB freed, 4337.46 MB remains loaded, 3.12 MB buffer reserved, lowvram patches: 0
2026-05-18T18:03:19.732772 - Unloaded partially: 171.88 MB freed, 4165.58 MB remains loaded, 3.12 MB buffer reserved, lowvram patches: 0
2026-05-18T18:03:19.735608 - 0 models unloaded.
2026-05-18T18:03:22.564634 - loaded completely; 159.56 MB loaded, full load: True
2026-05-18T18:03:33.117233 - Prompt executed in 00:24:56
2026-05-18T18:09:22.965271 - got prompt
2026-05-18T18:09:23.660587 - Prompt executed in 0.24 seconds
2026-05-18T18:09:53.535889 - got prompt
2026-05-18T18:10:27.456549 - loaded completely; 4897.05 MB loaded, full load: True
2026-05-18T18:11:23.451894 -
4%|▍ | 1/24 [00:53<20:34, 53.69s/it]2026-05-18T18:12:07.617310 - got prompt
2026-05-18T18:12:48.321146 - Interrupting prompt da1ee97f-e709-481b-ba8d-596c2d198f75
2026-05-18T18:12:49.147866 - Interrupting prompt da1ee97f-e709-481b-ba8d-596c2d198f75
2026-05-18T18:12:49.276856 -
4%|▍ | 1/24 [02:19<53:29, 139.53s/it]2026-05-18T18:12:49.277310 -
2026-05-18T18:12:49.317729 - Processing interrupted
2026-05-18T18:12:49.358411 - Prompt executed in 175.70 seconds
2026-05-18T18:12:52.787739 - Prompt da1ee97f-e709-481b-ba8d-596c2d198f75 is not currently running, skipping interrupt
2026-05-18T18:32:29.632070 -
100%|██████████| 24/24 [19:34<00:00, 50.39s/it]2026-05-18T18:32:29.637264 -
100%|██████████| 24/24 [19:34<00:00, 48.95s/it]2026-05-18T18:32:29.637316 -
2026-05-18T18:32:29.771989 - Requested to load AutoencoderKL
2026-05-18T18:32:43.795844 - Unloaded partially: 665.88 MB freed, 4231.21 MB remains loaded, 3.12 MB buffer reserved, lowvram patches: 0
2026-05-18T18:32:44.933277 - Unloaded partially: 81.25 MB freed, 4149.96 MB remains loaded, 3.12 MB buffer reserved, lowvram patches: 0
2026-05-18T18:32:44.934267 - 0 models unloaded.
2026-05-18T18:32:46.534864 - loaded completely; 159.56 MB loaded, full load: True
2026-05-18T18:32:59.110700 - Prompt executed in 00:20:05
2026-05-18T18:57:57.970009 - got prompt
2026-05-18T18:58:38.198286 - 0 models unloaded.
2026-05-18T18:59:26.177375 - Requested to load SDXL
2026-05-18T18:59:27.389831 - 0 models unloaded.
2026-05-18T19:00:05.149873 - loaded completely; 4897.05 MB loaded, full load: True
2026-05-18T20:23:35.489149 -
100%|██████████| 20/20 [1:23:29<00:00, 266.03s/it]2026-05-18T20:23:35.491795 -
100%|██████████| 20/20 [1:23:29<00:00, 250.46s/it]2026-05-18T20:23:35.491834 -
2026-05-18T20:23:37.391353 - Requested to load AutoencoderKL
2026-05-18T20:24:18.806543 - 0 models unloaded.
2026-05-18T20:24:28.575967 - loaded completely; 159.56 MB loaded, full load: True
2026-05-18T20:29:36.543939 - !!! Exception during processing !!! MPS backend out of memory (MPS allocated: 7.27 GiB, other allocations: 80.14 MiB, max allowed: 9.07 GiB). Tried to allocate 2.06 GiB on private pool. Use PYTORCH_MPS_HIGH_WATERMARK_RATIO=0.0 to disable upper limit for memory allocations (may cause system failure).
2026-05-18T20:29:37.130575 - Traceback (most recent call last):
File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/execution.py", line 535, 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 "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/execution.py", line 335, 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 "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/execution.py", line 309, in _async_map_node_over_list
await process_inputs(input_dict, i)
File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/execution.py", line 297, in process_inputs
result = f(**inputs)
^^^^^^^^^^^
File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/nodes.py", line 316, in decode
images = vae.decode(latent)
^^^^^^^^^^^^^^^^^^
File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/comfy/sd.py", line 1013, in decode
model_management.raise_non_oom(e)
File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/comfy/model_management.py", line 286, in raise_non_oom
raise e
File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/comfy/sd.py", line 1006, in decode
out = self.first_stage_model.decode(samples, **vae_options).to(device=self.output_device, dtype=self.vae_output_dtype(), copy=True)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/comfy/ldm/models/autoencoder.py", line 254, in decode
dec = self.decoder(dec, **decoder_kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1776, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1787, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/comfy/ldm/modules/diffusionmodules/model.py", line 808, in forward
h1 = self.up[i_level].block[i_block](h1, temb, conv_carry_in, conv_carry_out, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1776, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1787, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/comfy/ldm/modules/diffusionmodules/model.py", line 214, in forward
h = self.norm1(h)
^^^^^^^^^^^^^
File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1776, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1787, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/comfy/ops.py", line 563, in forward
return super().forward(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/nn/modules/normalization.py", line 327, in forward
return F.group_norm(input, self.num_groups, self.weight, self.bias, self.eps)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/nn/functional.py", line 2990, in group_norm
return torch.group_norm(
^^^^^^^^^^^^^^^^^
File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/_refs/__init__.py", line 3356, in native_group_norm
out = (input_reshaped - mean) * rstd
~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~
RuntimeError: MPS backend out of memory (MPS allocated: 7.27 GiB, other allocations: 80.14 MiB, max allowed: 9.07 GiB). Tried to allocate 2.06 GiB on private pool. Use PYTORCH_MPS_HIGH_WATERMARK_RATIO=0.0 to disable upper limit for memory allocations (may cause system failure).

2026-05-18T20:29:37.552873 - Prompt executed in 01:31:39
2026-05-18T20:33:05.722953 - got prompt
2026-05-18T20:33:07.351202 - Requested to load SDXL
2026-05-18T20:33:13.799027 - 0 models unloaded.
2026-05-18T20:33:53.382066 - loaded completely; 4897.05 MB loaded, full load: True
2026-05-18T21:44:52.081653 -
100%|██████████| 20/20 [1:10:56<00:00, 224.68s/it]2026-05-18T21:44:52.085267 -
100%|██████████| 20/20 [1:10:56<00:00, 212.84s/it]2026-05-18T21:44:52.085312 -
2026-05-18T21:44:53.096014 - Requested to load AutoencoderKL
2026-05-18T21:45:25.357707 - 0 models unloaded.
2026-05-18T21:45:33.199546 - loaded completely; 159.56 MB loaded, full load: True
2026-05-18T21:50:18.055672 - !!! Exception during processing !!! MPS backend out of memory (MPS allocated: 7.27 GiB, other allocations: 80.14 MiB, max allowed: 9.07 GiB). Tried to allocate 2.06 GiB on private pool. Use PYTORCH_MPS_HIGH_WATERMARK_RATIO=0.0 to disable upper limit for memory allocations (may cause system failure).
2026-05-18T21:50:18.390055 - Traceback (most recent call last):
File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/execution.py", line 535, 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 "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/execution.py", line 335, 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 "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/execution.py", line 309, in _async_map_node_over_list
await process_inputs(input_dict, i)
File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/execution.py", line 297, in process_inputs
result = f(**inputs)
^^^^^^^^^^^
File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/nodes.py", line 316, in decode
images = vae.decode(latent)
^^^^^^^^^^^^^^^^^^
File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/comfy/sd.py", line 1013, in decode
model_management.raise_non_oom(e)
File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/comfy/model_management.py", line 286, in raise_non_oom
raise e
File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/comfy/sd.py", line 1006, in decode
out = self.first_stage_model.decode(samples, **vae_options).to(device=self.output_device, dtype=self.vae_output_dtype(), copy=True)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/comfy/ldm/models/autoencoder.py", line 254, in decode
dec = self.decoder(dec, **decoder_kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1776, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1787, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/comfy/ldm/modules/diffusionmodules/model.py", line 808, in forward
h1 = self.up[i_level].block[i_block](h1, temb, conv_carry_in, conv_carry_out, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1776, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1787, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/comfy/ldm/modules/diffusionmodules/model.py", line 214, in forward
h = self.norm1(h)
^^^^^^^^^^^^^
File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1776, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1787, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/comfy/ops.py", line 563, in forward
return super().forward(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/nn/modules/normalization.py", line 327, in forward
return F.group_norm(input, self.num_groups, self.weight, self.bias, self.eps)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/nn/functional.py", line 2990, in group_norm
return torch.group_norm(
^^^^^^^^^^^^^^^^^
File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/_refs/__init__.py", line 3356, in native_group_norm
out = (input_reshaped - mean) * rstd
~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~
RuntimeError: MPS backend out of memory (MPS allocated: 7.27 GiB, other allocations: 80.14 MiB, max allowed: 9.07 GiB). Tried to allocate 2.06 GiB on private pool. Use PYTORCH_MPS_HIGH_WATERMARK_RATIO=0.0 to disable upper limit for memory allocations (may cause system failure).

2026-05-18T21:50:18.554796 - Prompt executed in 01:17:12
2026-05-19T11:49:29.793425 - got prompt
2026-05-19T11:49:31.529397 - Requested to load SDXL
2026-05-19T11:49:39.382972 - 0 models unloaded.
2026-05-19T11:50:17.536811 - loaded completely; 4897.05 MB loaded, full load: True
2026-05-19T13:16:45.722759 -
100%|██████████| 20/20 [1:26:26<00:00, 272.20s/it]2026-05-19T13:16:45.730881 -
100%|██████████| 20/20 [1:26:26<00:00, 259.33s/it]2026-05-19T13:16:45.730916 -
2026-05-19T13:16:54.275144 - Requested to load AutoencoderKL
2026-05-19T13:17:31.367006 - 0 models unloaded.
2026-05-19T13:17:38.579735 - loaded completely; 159.56 MB loaded, full load: True
2026-05-19T13:19:19.561247 - !!! Exception during processing !!! MPS backend out of memory (MPS allocated: 7.27 GiB, other allocations: 80.14 MiB, max allowed: 9.07 GiB). Tried to allocate 2.06 GiB on private pool. Use PYTORCH_MPS_HIGH_WATERMARK_RATIO=0.0 to disable upper limit for memory allocations (may cause system failure).
2026-05-19T13:19:19.760364 - Traceback (most recent call last):
File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/execution.py", line 535, 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 "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/execution.py", line 335, 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 "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/execution.py", line 309, in _async_map_node_over_list
await process_inputs(input_dict, i)
File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/execution.py", line 297, in process_inputs
result = f(**inputs)
^^^^^^^^^^^
File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/nodes.py", line 316, in decode
images = vae.decode(latent)
^^^^^^^^^^^^^^^^^^
File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/comfy/sd.py", line 1013, in decode
model_management.raise_non_oom(e)
File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/comfy/model_management.py", line 286, in raise_non_oom
raise e
File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/comfy/sd.py", line 1006, in decode
out = self.first_stage_model.decode(samples, **vae_options).to(device=self.output_device, dtype=self.vae_output_dtype(), copy=True)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/comfy/ldm/models/autoencoder.py", line 254, in decode
dec = self.decoder(dec, **decoder_kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1776, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1787, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/comfy/ldm/modules/diffusionmodules/model.py", line 808, in forward
h1 = self.up[i_level].block[i_block](h1, temb, conv_carry_in, conv_carry_out, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1776, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1787, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/comfy/ldm/modules/diffusionmodules/model.py", line 214, in forward
h = self.norm1(h)
^^^^^^^^^^^^^
File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1776, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1787, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Applications/ComfyUI.app/Contents/Resources/ComfyUI/comfy/ops.py", line 563, in forward
return super().forward(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/nn/modules/normalization.py", line 327, in forward
return F.group_norm(input, self.num_groups, self.weight, self.bias, self.eps)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/nn/functional.py", line 2990, in group_norm
return torch.group_norm(
^^^^^^^^^^^^^^^^^
File "/Volumes/一个张张盘/confyui/.venv/lib/python3.12/site-packages/torch/_refs/__init__.py", line 3356, in native_group_norm
out = (input_reshaped - mean) * rstd
~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~
RuntimeError: MPS backend out of memory (MPS allocated: 7.27 GiB, other allocations: 80.14 MiB, max allowed: 9.07 GiB). Tried to allocate 2.06 GiB on private pool. Use PYTORCH_MPS_HIGH_WATERMARK_RATIO=0.0 to disable upper limit for memory allocations (may cause system failure).

2026-05-19T13:19:19.847625 - Prompt executed in 01:29:49

```
## Attached Workflow
Please make sure that workflow does not contain any sensitive information such as API keys or passwords.
```
{"id":"dd305009-c7f3-4273-91f5-7d12755da764","revision":0,"last_node_id":16,"last_link_id":20,"nodes":[{"id":7,"type":"CLIPTextEncode","pos":[410,390],"size":[425.265625,180.59375],"flags":{},"order":4,"mode":0,"inputs":[{"localized_name":"clip","name":"clip","type":"CLIP","link":13},{"localized_name":"文本","name":"text","type":"STRING","widget":{"name":"text"},"link":null}],"outputs":[{"localized_name":"条件","name":"CONDITIONING","type":"CONDITIONING","slot_index":0,"links":[6]}],"properties":{"Node name for S&R":"CLIPTextEncode"},"widgets_values":["embedding：easynegative"],"color":"#223","bgcolor":"#335"},{"id":3,"type":"KSampler","pos":[860,190],"size":[315,317.328125],"flags":{},"order":7,"mode":0,"inputs":[{"localized_name":"模型","name":"model","type":"MODEL","link":16},{"localized_name":"正面条件","name":"positive","type":"CONDITIONING","link":4},{"localized_name":"负面条件","name":"negative","type":"CONDITIONING","link":6},{"localized_name":"Latent图像","name":"latent_image","type":"LATENT","link":18},{"localized_name":"种子","name":"seed","type":"INT","widget":{"name":"seed"},"link":null},{"localized_name":"步数","name":"steps","type":"INT","widget":{"name":"steps"},"link":null},{"localized_name":"cfg","name":"cfg","type":"FLOAT","widget":{"name":"cfg"},"link":null},{"localized_name":"采样器名称","name":"sampler_name","type":"COMBO","widget":{"name":"sampler_name"},"link":null},{"localized_name":"调度器","name":"scheduler","type":"COMBO","widget":{"name":"scheduler"},"link":null},{"localized_name":"降噪","name":"denoise","type":"FLOAT","widget":{"name":"denoise"},"link":null}],"outputs":[{"localized_name":"Latent","name":"LATENT","type":"LATENT","slot_index":0,"links":[7]}],"properties":{"Node name for S&R":"KSampler"},"widgets_values":[370569594667077,"randomize",20,8,"euler","normal",1]},{"id":8,"type":"VAEDecode","pos":[1210,190],"size":[225,100],"flags":{},"order":8,"mode":0,"inputs":[{"localized_name":"Latent","name":"samples","type":"LATENT","link":7},{"localized_name":"vae","name":"vae","type":"VAE","link":17}],"outputs":[{"localized_name":"图像","name":"IMAGE","type":"IMAGE","slot_index":0,"links":[9]}],"properties":{"Node name for S&R":"VAEDecode"},"widgets_values":[]},{"id":9,"type":"SaveImage","pos":[847.5615419295896,545.7596960388355],"size":[320.125,483.53125],"flags":{},"order":9,"mode":0,"inputs":[{"localized_name":"图片","name":"images","type":"IMAGE","link":9},{"localized_name":"文件名前缀","name":"filename_prefix","type":"STRING","widget":{"name":"filename_prefix"},"link":null}],"outputs":[],"properties":{"Node name for S&R":"SaveImage"},"widgets_values":["ComfyUI"]},{"id":6,"type":"CLIPTextEncode","pos":[170,-30],"size":[422.84375,164.3125],"flags":{},"order":5,"mode":0,"inputs":[{"localized_name":"clip","name":"clip","type":"CLIP","link":14},{"localized_name":"文本","name":"text","type":"STRING","widget":{"name":"text"},"link":null}],"outputs":[{"localized_name":"条件","name":"CONDITIONING","type":"CONDITIONING","slot_index":0,"links":[4]}],"properties":{"Node name for S&R":"CLIPTextEncode"},"widgets_values":["High quality, style drawing\nThe background color is white and the details are retained\n"]},{"id":11,"type":"CheckpointLoaderSimple","pos":[-320,390],"size":[270,130.65625],"flags":{},"order":0,"mode":0,"inputs":[{"localized_name":"Checkpoint名称","name":"ckpt_name","type":"COMBO","widget":{"name":"ckpt_name"},"link":null}],"outputs":[{"localized_name":"模型","name":"MODEL","type":"MODEL","links":[16]},{"localized_name":"CLIP","name":"CLIP","type":"CLIP","links":[13,14]},{"localized_name":"VAE","name":"VAE","type":"VAE","links":[17,19]}],"properties":{"Node name for S&R":"CheckpointLoaderSimple"},"widgets_values":["epicrealismXL_pureFix.safetensors"]},{"id":13,"type":"VAEEncode","pos":[200,600],"size":[225,72],"flags":{},"order":6,"mode":0,"inputs":[{"localized_name":"像素","name":"pixels","type":"IMAGE","link":20},{"localized_name":"vae","name":"vae","type":"VAE","link":19}],"outputs":[{"localized_name":"Latent","name":"LATENT","type":"LATENT","links":[18]}],"properties":{"Node name for S&R":"VAEEncode"},"widgets_values":[]},{"id":14,"type":"LoadImage","pos":[-300,620],"size":[270,344],"flags":{},"order":1,"mode":0,"inputs":[{"localized_name":"图像","name":"image","type":"COMBO","widget":{"name":"image"},"link":null},{"localized_name":"选择文件上传","name":"upload","type":"IMAGEUPLOAD","widget":{"name":"upload"},"link":null}],"outputs":[{"localized_name":"图像","name":"IMAGE","type":"IMAGE","links":[20]},{"localized_name":"遮罩","name":"MASK","type":"MASK","links":null}],"properties":{"Node name for S&R":"LoadImage"},"widgets_values":["d81bb5e2fb41a4cc75bc52328dcda4ba.jpg","image"]},{"id":15,"type":"LoadImage","pos":[176265.16701772975,407.8345762334375],"size":[270,340],"flags":{},"order":2,"mode":0,"inputs":[{"localized_name":"图像","name":"image","type":"COMBO","widget":{"name":"image"},"link":null},{"localized_name":"选择文件上传","name":"upload","type":"IMAGEUPLOAD","widget":{"name":"upload"},"link":null}],"outputs":[{"localized_name":"图像","name":"IMAGE","type":"IMAGE","links":null},{"localized_name":"遮罩","name":"MASK","type":"MASK","links":null}],"properties":{"Node name for S&R":"LoadImage"},"widgets_values":["39ccfe4938004a76b2642f6bc158087e.jpeg~tplv-a9rns2rl98-image_pre_watermark_1_5b.png","image"]},{"id":16,"type":"PreviewImage","pos":[-441.3007089526918,-1153.824459332639],"size":[225,48],"flags":{},"order":3,"mode":0,"inputs":[{"localized_name":"图像","name":"images","type":"IMAGE","link":null}],"outputs":[],"properties":{"Node name for S&R":"PreviewImage"},"widgets_values":[]}],"links":[[4,6,0,3,1,"CONDITIONING"],[6,7,0,3,2,"CONDITIONING"],[7,3,0,8,0,"LATENT"],[9,8,0,9,0,"IMAGE"],[13,11,1,7,0,"CLIP"],[14,11,1,6,0,"CLIP"],[16,11,0,3,0,"MODEL"],[17,11,2,8,1,"VAE"],[18,13,0,3,3,"LATENT"],[19,11,2,13,1,"VAE"],[20,14,0,13,0,"IMAGE"]],"groups":[],"config":{},"extra":{"ds":{"scale":0.1,"offset":[7590.011646452692,5343.551021832639]},"linearData":{"inputs":[],"outputs":[]},"frontendVersion":"1.43.18"},"version":0.4}
```

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