Import errors appear when launching ComfyUI

[cedico]

Hi,

I’ve installed ComfyUI and it works fine to generate an image.
but when launching I can see some import errors:

WARNING: some comfy_extras/ nodes did not import correctly. This may be because they are missing some dependencies.
-IMPORT FAILED: nodes_canny.py
-IMPORT FAILED: nodes_morphology.py
-IMPORT FAILED: nodes_audio.py
-IMPORT FAILED: nodes_video.py
=> This issue might be caused by new missing dependencies added the last time you updated ComfyUI.
Please do a: pip install -r requirements-ipex.txt

Could you tell me how to correct the issue please ?

[ltdrdata]

Please upload the full log.

[cedico]

`(comfyui_env) F:\AI\VideoComfy\ComfyUI>python main.py --bf16-unet
[START] Security scan
[DONE] Security scan

ComfyUI-Manager: installing dependencies done.

** ComfyUI startup time: 2025-02-24 10:20:45.125
** Platform: Windows
** Python version: 3.10.11 (tags/v3.10.11:7d4cc5a, Apr 5 2023, 00:38:17) [MSC v.1929 64 bit (AMD64)]
** Python executable: F:\AI\VideoComfy\ComfyUI\comfyui_env\Scripts\python.exe
** ComfyUI Path: F:\AI\VideoComfy\ComfyUI
** ComfyUI Base Folder Path: F:\AI\VideoComfy\ComfyUI
** User directory: F:\AI\VideoComfy\ComfyUI\user
** ComfyUI-Manager config path: F:\AI\VideoComfy\ComfyUI\user\default\ComfyUI-Manager\config.ini
** Log path: F:\AI\VideoComfy\ComfyUI\user\comfyui.log
[ComfyUI-Manager] PyTorch is not installed

Prestartup times for custom nodes:
5.4 seconds: F:\AI\VideoComfy\ComfyUI\custom_nodes\ComfyUI-Manager

Warning, you are using an old pytorch version and some ckpt/pt files might be loaded unsafely. Upgrading to 2.4 or above is recommended.
F:\AI\VideoComfy\ComfyUI\comfyui_env\lib\site-packages\torchvision\io\image.py:13: UserWarning: Failed to load image Python extension: ''If you don't plan on using image functionality from torchvision.io, you can ignore this warning. Otherwise, there might be something wrong with your environment. Did you have libjpeg or libpng installed before building torchvision from source?
warn(
No CUDA runtime is found, using CUDA_HOME='C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA\v11.8'
ZE_LOADER_DEBUG_TRACE:Using Loader Library Path:
ZE_LOADER_DEBUG_TRACE:Tracing Layer Library Path: ze_tracing_layer.dll
Total VRAM 15931 MB, total RAM 65306 MB
pytorch version: 2.1.0a0+cxx11.abi
Set vram state to: NORMAL_VRAM
Device: xpu
Using pytorch attention
ComfyUI version: 0.3.14
[Prompt Server] web root: F:\AI\VideoComfy\ComfyUI\web
Traceback (most recent call last):
File "F:\AI\VideoComfy\ComfyUI\nodes.py", line 2145, in load_custom_node
module_spec.loader.exec_module(module)
File "", line 883, in exec_module
File "", line 241, in _call_with_frames_removed
File "F:\AI\VideoComfy\ComfyUI\comfy_extras\nodes_canny.py", line 1, in
from kornia.filters import canny
ModuleNotFoundError: No module named 'kornia'

Cannot import F:\AI\VideoComfy\ComfyUI\comfy_extras\nodes_canny.py module for custom nodes: No module named 'kornia'
Traceback (most recent call last):
File "F:\AI\VideoComfy\ComfyUI\nodes.py", line 2145, in load_custom_node
module_spec.loader.exec_module(module)
File "", line 883, in exec_module
File "", line 241, in _call_with_frames_removed
File "F:\AI\VideoComfy\ComfyUI\comfy_extras\nodes_morphology.py", line 4, in
from kornia.morphology import dilation, erosion, opening, closing, gradient, top_hat, bottom_hat
ModuleNotFoundError: No module named 'kornia'

Cannot import F:\AI\VideoComfy\ComfyUI\comfy_extras\nodes_morphology.py module for custom nodes: No module named 'kornia'
Traceback (most recent call last):
File "F:\AI\VideoComfy\ComfyUI\nodes.py", line 2145, in load_custom_node
module_spec.loader.exec_module(module)
File "", line 883, in exec_module
File "", line 241, in _call_with_frames_removed
File "F:\AI\VideoComfy\ComfyUI\comfy_extras\nodes_audio.py", line 1, in
import torchaudio
ModuleNotFoundError: No module named 'torchaudio'

Cannot import F:\AI\VideoComfy\ComfyUI\comfy_extras\nodes_audio.py module for custom nodes: No module named 'torchaudio'
Traceback (most recent call last):
File "F:\AI\VideoComfy\ComfyUI\nodes.py", line 2145, in load_custom_node
module_spec.loader.exec_module(module)
File "", line 883, in exec_module
File "", line 241, in _call_with_frames_removed
File "F:\AI\VideoComfy\ComfyUI\comfy_extras\nodes_video.py", line 2, in
import av
ModuleNotFoundError: No module named 'av'

Cannot import F:\AI\VideoComfy\ComfyUI\comfy_extras\nodes_video.py module for custom nodes: No module named 'av'

Loading: ComfyUI-Impact-Pack (V8.8.1)

[Impact Pack] Wildcards loading done.

Loading: ComfyUI-Manager (V3.25.1)

[ComfyUI-Manager] network_mode: public

ComfyUI Version: v0.3.14-39-gb4d3652d | Released on '2025-02-19'

[comfyui_controlnet_aux] | INFO -> Using ckpts path: F:\AI\VideoComfy\ComfyUI\custom_nodes\comfyui_controlnet_aux\ckpts
[comfyui_controlnet_aux] | INFO -> Using symlinks: False
[comfyui_controlnet_aux] | INFO -> Using ort providers: ['CUDAExecutionProvider', 'DirectMLExecutionProvider', 'OpenVINOExecutionProvider', 'ROCMExecutionProvider', 'CPUExecutionProvider', 'CoreMLExecutionProvider']
[ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/github-stats.json
[ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/extension-node-map.json
[ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/custom-node-list.json
[ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/model-list.json
[ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/alter-list.json
F:\AI\VideoComfy\ComfyUI\custom_nodes\comfyui_controlnet_aux\node_wrappers\dwpose.py:26: UserWarning: DWPose: Onnxruntime not found or doesn't come with acceleration providers, switch to OpenCV with CPU device. DWPose might run very slowly
warnings.warn("DWPose: Onnxruntime not found or doesn't come with acceleration providers, switch to OpenCV with CPU device. DWPose might run very slowly")
Efficiency Nodes: Attempting to add Control Net options to the 'HiRes-Fix Script' Node (comfyui_controlnet_aux add-on)...Success!

Import times for custom nodes:
0.0 seconds: F:\AI\VideoComfy\ComfyUI\custom_nodes\websocket_image_save.py
0.5 seconds: F:\AI\VideoComfy\ComfyUI\custom_nodes\efficiency-nodes-comfyui
0.8 seconds: F:\AI\VideoComfy\ComfyUI\custom_nodes\ComfyUI-Manager
1.8 seconds: F:\AI\VideoComfy\ComfyUI\custom_nodes\comfyui-impact-pack
2.4 seconds: F:\AI\VideoComfy\ComfyUI\custom_nodes\comfyui_controlnet_aux

WARNING: some comfy_extras/ nodes did not import correctly. This may be because they are missing some dependencies.

IMPORT FAILED: nodes_canny.py
IMPORT FAILED: nodes_morphology.py
IMPORT FAILED: nodes_audio.py
IMPORT FAILED: nodes_video.py

This issue might be caused by new missing dependencies added the last time you updated ComfyUI.
Please do a: pip install -r requirements.txt

Starting server

To see the GUI go to: http://127.0.0.1:8188
FETCH ComfyRegistry Data: 5/35
FETCH ComfyRegistry Data: 10/35
FETCH ComfyRegistry Data: 15/35
FETCH ComfyRegistry Data: 20/35
FETCH ComfyRegistry Data: 25/35
FETCH ComfyRegistry Data: 30/35
FETCH ComfyRegistry Data: 35/35
FETCH ComfyRegistry Data [DONE]
[ComfyUI-Manager] default cache updated: https://api.comfy.org/nodes
nightly_channel: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/remote
FETCH DATA from: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/custom-node-list.json [DONE]
[ComfyUI-Manager] All startup tasks have been completed.
got prompt
model weight dtype torch.bfloat16, manual cast: None
model_type EPS`

[ltdrdata]

How did you install the ComfyUI?

[cedico]

I used the basic way I think:

git clone /~https://github.com/comfyanonymous/ComfyUI
python -m venv comfyui_env
comfyui_env\scripts\activate
pip install -r requirements-ipex.txt

and RUN : python main.py --bf16-unet 

[ltdrdata]

/~https://github.com/comfyanonymous/ComfyUI?tab=readme-ov-file#manual-install-windows-linux

[cedico]

I've already tried this way but I met some other errors, and was not able to generate any image.
But I am going to try again and send you my error logs
Thank you for your help

[cedico]

You would advise me to install it which OPTION 1 or 2? (I have an Intel ARC A770 GPU)

Intel GPUs (Windows and Linux)

> (Option 1) Intel Arc GPU users can install native PyTorch with torch.xpu support using pip (currently available in PyTorch nightly builds). More information can be found here

To install PyTorch nightly, use the following command:
pip install --pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/xpu

Launch ComfyUI by running python main.py

> (Option 2) Alternatively, Intel GPUs supported by Intel Extension for PyTorch (IPEX) can leverage IPEX for improved performance.

For Intel® Arc™ A-Series Graphics utilizing IPEX, create a conda environment and use the commands below:
conda install libuv
pip install torch==2.3.1.post0+cxx11.abi torchvision==0.18.1.post0+cxx11.abi torchaudio==2.3.1.post0+cxx11.abi intel-extension-for-pytorch==2.3.110.post0+xpu --extra-index-url https://pytorch-extension.intel.com/release-whl/stable/xpu/us/ --extra-index-url https://pytorch-extension.intel.com/release-whl/stable/xpu/cn/
For other supported Intel GPUs with IPEX, visit Installation for more information.

Additional discussion and help can be found #476.

[cedico]

Hi,
I did this (OPTION 2), but I at the end, when launching main.py, I have a lot of missing libs (yaml, safetensors...) :

- Git clone /~https://github.com/comfyanonymous/ComfyUI
- conda create -n ComfyUI python=3.11 libuv
- conda activate ComfyUI
- pip install torch==2.3.1.post0+cxx11.abi torchvision==0.18.1.post0+cxx11.abi torchaudio==2.3.1.post0+cxx11.abi intel-extension-for-pytorch==2.3.110.post0+xpu --extra-index-url https://pytorch-extension.intel.com/release-whl/stable/xpu/us/ --extra-index-url https://pytorch-extension.intel.com/release-whl/stable/xpu/cn/
- python main.py

Am I doing something wrong? (on your link, they said to create an environment first and then to install libuv)

[ltdrdata]

As mentioned in the thread you need to switch to Linux if you chose option 2.

[cedico]

I've just started with the OPTION 1 for Windows:

cd C:\AI
git clone /~https://github.com/comfyanonymous/ComfyUI
cd C:\AI\ComfyUI
copy checkpoints and VAE files...
pip install --pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/xpu
python main.py

But now I have this error:
"ModuleNotFoundError: No module named 'yaml'"
and if I try to install this module, I have another one missing, etc...

not that easy to follow :(

[ltdrdata]

I've just started with the OPTION 1 for Windows:

cd C:\AI
git clone /~https://github.com/comfyanonymous/ComfyUI
cd C:\AI\ComfyUI
copy checkpoints and VAE files...
pip install --pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/xpu
python main.py

But now I have this error:
"ModuleNotFoundError: No module named 'yaml'"
and if I try to install this module, I have another one missing, etc...

not that easy to follow :(

That is a separated issue. You need install requirements.txt in the ComfyUI folder.

[ltdrdata]

I've just started with the OPTION 1 for Windows:

cd C:\AI
git clone /~https://github.com/comfyanonymous/ComfyUI
cd C:\AI\ComfyUI
copy checkpoints and VAE files...
pip install --pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/xpu
python main.py

But now I have this error:
"ModuleNotFoundError: No module named 'yaml'"
and if I try to install this module, I have another one missing, etc...

not that easy to follow :(

That is a separated issue. You need install requirements.txt in the ComfyUI folder.

[cedico]

Simple question: Why install "requirements.txt" and not "requirements-ipex.txt" ?
But this works fine to generate a simple image, thanks a lot ! :)

Now I have another issue. I'm testing a workflow with KSampler (Efficient) and I obtain this error :
KSampler (Efficient) Current platform can NOT allocate memory block with size larger than 4GB! Tried to allocate 8.00 GiB (GPU 0; 15.56 GiB total capacity; 2.93 GiB already allocated; 4.49 GiB reserved in total by PyTorch)

Could you tell me how to bypass this error or how to use a workaround ? reading the web, I read there are a lot of people who has the same error

[cedico]

Hi,
I've tested again, and I have a slightly different error :

KSampler (Efficient) : XPU out of memory. Tried to allocate 8.00 GiB. GPU 0 has a total capacity of 15.56 GiB. Of the allocated memory 2.93 GiB is allocated by PyTorch, and 418.94 MiB is reserved by PyTorch but unallocated. Please use empty_cache to release all unoccupied cached memory.

Can this error be counteracted by this empty_cache option? if yes, how to use it please?

Log details:
Efficient Loader Models Cache:
Ckpt:
[1] DreamShaper_8_pruned
Requested to load BaseModel
loaded completely 14268.33369140625 1639.406135559082 True
100%|██████████████████████████████████████████████████████████████████████████████████| 20/20 [00:09<00:00, 2.05it/s]
Requested to load AutoencoderKL
loaded completely 10985.40673828125 159.55708122253418 True
Prompt executed in 15.31 seconds
got prompt
Prompt executed in 0.00 seconds
got prompt
Requested to load ControlNet
loaded completely 12455.57490234375 689.0852355957031 True
0%| | 0/20 [00:00<?, ?it/s]
!!! Exception during processing !!! XPU out of memory. Tried to allocate 8.00 GiB. GPU 0 has a total capacity of 15.56 GiB. Of the allocated memory 2.93 GiB is allocated by PyTorch, and 418.94 MiB is reserved by PyTorch but unallocated. Please use empty_cache to release all unoccupied cached memory.
Traceback (most recent call last):
File "F:\AI\VideoComfy\ComfyUI\execution.py", line 327, in execute
output_data, output_ui, has_subgraph = get_output_data(obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
File "F:\AI\VideoComfy\ComfyUI\execution.py", line 202, in get_output_data
return_values = _map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
File "F:\AI\VideoComfy\ComfyUI\execution.py", line 174, in _map_node_over_list
process_inputs(input_dict, i)
File "F:\AI\VideoComfy\ComfyUI\execution.py", line 163, in process_inputs
results.append(getattr(obj, func)(**inputs))
File "F:\AI\VideoComfy\ComfyUI\custom_nodes\efficiency-nodes-comfyui\efficiency_nodes.py", line 732, in sample
samples, images, gifs, preview = process_latent_image(model, seed, steps, cfg, sampler_name, scheduler,
File "F:\AI\VideoComfy\ComfyUI\custom_nodes\efficiency-nodes-comfyui\efficiency_nodes.py", line 550, in process_latent_image
samples = KSampler().sample(model, seed, steps, cfg, sampler_name, scheduler, positive, negative,
File "F:\AI\VideoComfy\ComfyUI\nodes.py", line 1540, in sample
return common_ksampler(model, seed, steps, cfg, sampler_name, scheduler, positive, negative, latent_image, denoise=denoise)
File "F:\AI\VideoComfy\ComfyUI\nodes.py", line 1507, in common_ksampler
samples = comfy.sample.sample(model, noise, steps, cfg, sampler_name, scheduler, positive, negative, latent_image,
File "F:\AI\VideoComfy\ComfyUI\custom_nodes\comfyui-impact-pack\modules\impact\sample_error_enhancer.py", line 22, in informative_sample
raise e
File "F:\AI\VideoComfy\ComfyUI\custom_nodes\comfyui-impact-pack\modules\impact\sample_error_enhancer.py", line 9, in informative_sample
return original_sample(*args, **kwargs) # This code helps interpret error messages that occur within exceptions but does not have any impact on other operations.
File "F:\AI\VideoComfy\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 "F:\AI\VideoComfy\ComfyUI\comfy\samplers.py", line 1110, 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 "F:\AI\VideoComfy\ComfyUI\comfy\samplers.py", line 1000, in sample
return cfg_guider.sample(noise, latent_image, sampler, sigmas, denoise_mask, callback, disable_pbar, seed)
File "F:\AI\VideoComfy\ComfyUI\comfy\samplers.py", line 985, in sample
output = executor.execute(noise, latent_image, sampler, sigmas, denoise_mask, callback, disable_pbar, seed)
File "F:\AI\VideoComfy\ComfyUI\comfy\patcher_extension.py", line 110, in execute
return self.original(*args, **kwargs)
File "F:\AI\VideoComfy\ComfyUI\comfy\samplers.py", line 953, in outer_sample
output = self.inner_sample(noise, latent_image, device, sampler, sigmas, denoise_mask, callback, disable_pbar, seed)
File "F:\AI\VideoComfy\ComfyUI\comfy\samplers.py", line 936, in inner_sample
samples = executor.execute(self, sigmas, extra_args, callback, noise, latent_image, denoise_mask, disable_pbar)
File "F:\AI\VideoComfy\ComfyUI\comfy\patcher_extension.py", line 110, in execute
return self.original(*args, **kwargs)
File "F:\AI\VideoComfy\ComfyUI\comfy\samplers.py", line 715, 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:\Python310\lib\site-packages\torch\utils_contextlib.py", line 116, in decorate_context
return func(*args, **kwargs)
File "F:\AI\VideoComfy\ComfyUI\comfy\k_diffusion\sampling.py", line 161, in sample_euler
denoised = model(x, sigma_hat * s_in, **extra_args)
File "F:\AI\VideoComfy\ComfyUI\comfy\samplers.py", line 379, in call
out = self.inner_model(x, sigma, model_options=model_options, seed=seed)
File "F:\AI\VideoComfy\ComfyUI\comfy\samplers.py", line 916, in call
return self.predict_noise(*args, **kwargs)
File "F:\AI\VideoComfy\ComfyUI\comfy\samplers.py", line 919, 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 "F:\AI\VideoComfy\ComfyUI\comfy\samplers.py", line 359, in sampling_function
out = calc_cond_batch(model, conds, x, timestep, model_options)
File "F:\AI\VideoComfy\ComfyUI\comfy\samplers.py", line 195, in calc_cond_batch
return executor.execute(model, conds, x_in, timestep, model_options)
File "F:\AI\VideoComfy\ComfyUI\comfy\patcher_extension.py", line 110, in execute
return self.original(*args, **kwargs)
File "F:\AI\VideoComfy\ComfyUI\comfy\samplers.py", line 303, in calc_cond_batch
c['control'] = control.get_control(input_x, timestep, c, len(cond_or_uncond), transformer_options)
File "F:\AI\VideoComfy\ComfyUI\comfy\controlnet.py", line 273, in get_control
control = self.control_model(x=x_noisy.to(dtype), hint=self.cond_hint, timesteps=timestep.to(dtype), context=context.to(dtype), **extra)
File "C:\Python310\lib\site-packages\torch\nn\modules\module.py", line 1751, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "C:\Python310\lib\site-packages\torch\nn\modules\module.py", line 1762, in _call_impl
return forward_call(*args, **kwargs)
File "F:\AI\VideoComfy\ComfyUI\comfy\cldm\cldm.py", line 426, in forward
h = module(h, emb, context)
File "C:\Python310\lib\site-packages\torch\nn\modules\module.py", line 1751, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "C:\Python310\lib\site-packages\torch\nn\modules\module.py", line 1762, in _call_impl
return forward_call(*args, **kwargs)
File "F:\AI\VideoComfy\ComfyUI\comfy\ldm\modules\diffusionmodules\openaimodel.py", line 69, in forward
return forward_timestep_embed(self, *args, **kwargs)
File "F:\AI\VideoComfy\ComfyUI\comfy\ldm\modules\diffusionmodules\openaimodel.py", line 44, in forward_timestep_embed
x = layer(x, context, transformer_options)
File "C:\Python310\lib\site-packages\torch\nn\modules\module.py", line 1751, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "C:\Python310\lib\site-packages\torch\nn\modules\module.py", line 1762, in _call_impl
return forward_call(*args, **kwargs)
File "F:\AI\VideoComfy\ComfyUI\comfy\ldm\modules\attention.py", line 782, in forward
x = block(x, context=context[i], transformer_options=transformer_options)
File "C:\Python310\lib\site-packages\torch\nn\modules\module.py", line 1751, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "C:\Python310\lib\site-packages\torch\nn\modules\module.py", line 1762, in _call_impl
return forward_call(*args, **kwargs)
File "F:\AI\VideoComfy\ComfyUI\comfy\ldm\modules\attention.py", line 669, in forward
n = self.attn1(n, context=context_attn1, value=value_attn1)
File "C:\Python310\lib\site-packages\torch\nn\modules\module.py", line 1751, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "C:\Python310\lib\site-packages\torch\nn\modules\module.py", line 1762, in _call_impl
return forward_call(*args, **kwargs)
File "F:\AI\VideoComfy\ComfyUI\comfy\ldm\modules\attention.py", line 563, in forward
out = optimized_attention(q, k, v, self.heads, attn_precision=self.attn_precision)
File "F:\AI\VideoComfy\ComfyUI\comfy\ldm\modules\attention.py", line 441, in attention_pytorch
out = torch.nn.functional.scaled_dot_product_attention(q, k, v, attn_mask=mask, dropout_p=0.0, is_causal=False)
torch.OutOfMemoryError: XPU out of memory. Tried to allocate 8.00 GiB. GPU 0 has a total capacity of 15.56 GiB. Of the allocated memory 2.93 GiB is allocated by PyTorch, and 418.94 MiB is reserved by PyTorch but unallocated. Please use empty_cache to release all unoccupied cached memory.

Got an OOM, unloading all loaded models.
Prompt executed in 9.27 seconds