Closed divyjot closed 5 months ago
When run_realistc.bat is executed without any parameters to entry_with_update.py, fooocus used RAM to generate image and GPU usagew is 0%.
This is expected, please wait until your Laptop has loaded and moved all necessary models to generate the image into swap as well as RAM. You can check the hardware usag for RAM / Drivee in your task manager.
Please remove all additional arguments and try again. It may take some time but somewhen is going to generate the image, even if very slow.
Checklist
What happened?
When run_realistc.bat is executed without any parameters to entry_with_update.py, fooocus used RAM to generate image and GPU usagew is 0%. I have Asus laptop with NVIDIA gtx 1660 ti 6GB.
When, parameters like --always-gpu used, still does not use GPU. when added --disable-offload-from-vram it gives cuda out of memory. Output error below: `D:\Fooocus_win64_2-1-831>.\python_embeded\python.exe -s Fooocus\entry_with_update.py --preset realistic --always-gpu --disable-offload-from-vram Already up-to-date Update succeeded. [System ARGV] ['Fooocus\entry_with_update.py', '--preset', 'realistic', '--always-gpu', '--disable-offload-from-vram'] Python 3.10.9 (tags/v3.10.9:1dd9be6, Dec 6 2022, 20:01:21) [MSC v.1934 64 bit (AMD64)] Fooocus version: 2.3.1 Loaded preset: D:\Fooocus_win64_2-1-831\Fooocus\presets\realistic.json [Cleanup] Attempting to delete content of temp dir C:\Users\xxxxx\AppData\Local\Temp\fooocus [Cleanup] Cleanup successful Total VRAM 6144 MB, total RAM 15790 MB Set vram state to: HIGH_VRAM Device: cuda:0 NVIDIA GeForce GTX 1660 Ti : native VAE dtype: torch.float32 Using pytorch cross attention Refiner unloaded. Running on local URL: http://127.0.0.1:7865
To create a public link, set
share=True
inlaunch()
. IMPORTANT: You are using gradio version 3.41.2, however version 4.29.0 is available, please upgrade.model_type EPS UNet ADM Dimension 2816 Using pytorch attention in VAE Working with z of shape (1, 4, 32, 32) = 4096 dimensions. Using pytorch attention in VAE extra {'cond_stage_model.clip_l.text_projection', 'cond_stage_model.clip_g.transformer.text_model.embeddings.position_ids', 'cond_stage_model.clip_l.logit_scale'} loaded straight to GPU Requested to load SDXL Loading 1 new model [Fooocus Model Management] Moving model(s) has taken 0.16 seconds Base model loaded: D:\Fooocus_win64_2-1-831\Fooocus\models\checkpoints\realisticStockPhoto_v20.safetensors Request to load LoRAs [['SDXL_FILM_PHOTOGRAPHY_STYLE_BetaV0.4.safetensors', 0.25], ['None', 1.0], ['None', 1.0], ['None', 1.0], ['None', 1.0]] for model [D:\Fooocus_win64_2-1-831\Fooocus\models\checkpoints\realisticStockPhoto_v20.safetensors]. Loaded LoRA [D:\Fooocus_win64_2-1-831\Fooocus\models\loras\SDXL_FILM_PHOTOGRAPHY_STYLE_BetaV0.4.safetensors] for UNet [D:\Fooocus_win64_2-1-831\Fooocus\models\checkpoints\realisticStockPhoto_v20.safetensors] with 788 keys at weight 0.25. Loaded LoRA [D:\Fooocus_win64_2-1-831\Fooocus\models\loras\SDXL_FILM_PHOTOGRAPHY_STYLE_BetaV0.4.safetensors] for CLIP [D:\Fooocus_win64_2-1-831\Fooocus\models\checkpoints\realisticStockPhoto_v20.safetensors] with 264 keys at weight 0.25. Fooocus V2 Expansion: Vocab with 642 words. Fooocus Expansion engine loaded for cuda:0, use_fp16 = False. Requested to load SDXLClipModel Requested to load GPT2LMHeadModel Loading 2 new models ERROR clip_g.transformer.text_model.encoder.layers.12.self_attn.out_proj.weight CUDA out of memory. Tried to allocate 20.00 MiB. GPU 0 has a total capacty of 6.00 GiB of which 0 bytes is free. Of the allocated memory 12.25 GiB is allocated by PyTorch, and 230.03 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF Exception in thread Thread-2 (worker): Traceback (most recent call last): File "D:\Fooocus_win64_2-1-831\Fooocus\ldm_patched\modules\model_management.py", line 300, in model_load self.real_model = self.model.patch_model(device_to=patch_model_to) #TODO: do something with loras and offloading to CPU File "D:\Fooocus_win64_2-1-831\Fooocus\ldm_patched\modules\model_patcher.py", line 196, in patch_model self.backup[key] = weight.to(device=self.offload_device, copy=inplace_update) torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 26.00 MiB. GPU 0 has a total capacty of 6.00 GiB of which 0 bytes is free. Of the allocated memory 12.25 GiB is allocated by PyTorch, and 230.34 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
During handling of the above exception, another exception occurred:
Traceback (most recent call last): File "threading.py", line 1016, in _bootstrap_inner File "threading.py", line 953, in run File "D:\Fooocus_win64_2-1-831\Fooocus\modules\async_worker.py", line 32, in worker import modules.default_pipeline as pipeline File "D:\Fooocus_win64_2-1-831\Fooocus\modules\default_pipeline.py", line 254, in
refresh_everything(
File "D:\Fooocus_win64_2-1-831\python_embeded\lib\site-packages\torch\utils_contextlib.py", line 115, in decorate_context
return func(*args, kwargs)
File "D:\Fooocus_win64_2-1-831\python_embeded\lib\site-packages\torch\utils_contextlib.py", line 115, in decorate_context
return func(*args, *kwargs)
File "D:\Fooocus_win64_2-1-831\Fooocus\modules\default_pipeline.py", line 249, in refresh_everything
prepare_text_encoder(async_call=True)
File "D:\Fooocus_win64_2-1-831\python_embeded\lib\site-packages\torch\utils_contextlib.py", line 115, in decorate_context
return func(args, kwargs)
File "D:\Fooocus_win64_2-1-831\python_embeded\lib\site-packages\torch\utils_contextlib.py", line 115, in decorate_context
return func(*args, *kwargs)
File "D:\Fooocus_win64_2-1-831\Fooocus\modules\default_pipeline.py", line 212, in prepare_text_encoder
ldm_patched.modules.model_management.load_models_gpu([final_clip.patcher, final_expansion.patcher])
File "D:\Fooocus_win64_2-1-831\Fooocus\modules\patch.py", line 447, in patched_load_models_gpu
y = ldm_patched.modules.model_management.load_models_gpu_origin(args, **kwargs)
File "D:\Fooocus_win64_2-1-831\Fooocus\ldm_patched\modules\model_management.py", line 437, in load_models_gpu
cur_loaded_model = loaded_model.model_load(lowvram_model_memory)
File "D:\Fooocus_win64_2-1-831\Fooocus\ldm_patched\modules\model_management.py", line 302, in model_load
self.model.unpatch_model(self.model.offload_device)
File "D:\Fooocus_win64_2-1-831\Fooocus\ldm_patched\modules\model_patcher.py", line 350, in unpatch_model
self.model.to(device_to)
File "D:\Fooocus_win64_2-1-831\python_embeded\lib\site-packages\torch\nn\modules\module.py", line 1160, in to
return self._apply(convert)
File "D:\Fooocus_win64_2-1-831\python_embeded\lib\site-packages\torch\nn\modules\module.py", line 810, in _apply
module._apply(fn)
File "D:\Fooocus_win64_2-1-831\python_embeded\lib\site-packages\torch\nn\modules\module.py", line 810, in _apply
module._apply(fn)
File "D:\Fooocus_win64_2-1-831\python_embeded\lib\site-packages\torch\nn\modules\module.py", line 810, in _apply
module._apply(fn)
[Previous line repeated 5 more times]
File "D:\Fooocus_win64_2-1-831\python_embeded\lib\site-packages\torch\nn\modules\module.py", line 833, in _apply
param_applied = fn(param)
File "D:\Fooocus_win64_2-1-831\python_embeded\lib\site-packages\torch\nn\modules\module.py", line 1158, in convert
return t.to(device, dtype if t.is_floating_point() or t.is_complex() else None, non_blocking)
torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 26.00 MiB. GPU 0 has a total capacty of 6.00 GiB of which 0 bytes is free. Of the allocated memory 11.76 GiB is allocated by PyTorch, and 731.59 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF`
Steps to reproduce the problem
execute run_realistic.bat file. give prompt
What should have happened?
Used the GPU instead of RAM and CPU
What browsers do you use to access Fooocus?
Mozilla Firefox
Where are you running Fooocus?
Locally
What operating system are you using?
Windows 11
Console logs
Additional information
everything is up to date