Open jk779 opened 5 months ago
what happens if you don't put "--force-fp16" ?
it seems to not make any differnce unfortunately :/
venv/bin/python main.py
Total VRAM 65536 MB, total RAM 65536 MB
Set vram state to: SHARED
Device: mps
VAE dtype: torch.float32
Using sub quadratic optimization for cross attention, if you have memory or speed issues try using: --use-split-cross-attention
Import times for custom nodes:
0.0 seconds: /Users/michael/Projects/ai/ComfyUI/custom_nodes/websocket_image_save.py
Starting server
To see the GUI go to: http://127.0.0.1:8188
got prompt
model_type STABLE_CASCADE
clip missing: ['clip_g.logit_scale']
model_type STABLE_CASCADE
Missing VAE keys ['encoder.mean', 'encoder.std']
clip missing: ['clip_g.logit_scale']
Requested to load StableCascadeClipModel
Loading 1 new model
Requested to load StableCascade_C
Loading 1 new model
100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 20/20 [00:38<00:00, 1.92s/it]
Requested to load StableCascade_B
Loading 1 new model
100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 10/10 [00:25<00:00, 2.51s/it]
Requested to load StageA
Loading 1 new model
Prompt executed in 86.38 seconds
a slimilar image appears when not using the pre-version (2.4.0-pre) of pytorch, but the current one (2.2.2):
I did a seemingly full reinstall
~/Projects/ai/ComfyUI master*
rm -rf venv
~/Projects/ai/ComfyUI master*
❯ python3 -m venv venv
~/Projects/ai/ComfyUI master*
❯ source venv/bin/activate
~/Projects/ai/ComfyUI master*
venv ❯ venv/bin/pip --require-virtualenv --python venv/bin/python --no-cache-dir install -r requirements.txt
Collecting torch (from -r requirements.txt (line 1))
Downloading torch-2.2.2-cp312-none-macosx_11_0_arm64.whl.metadata (25 kB)
Collecting torchsde (from -r requirements.txt (line 2))
Downloading torchsde-0.2.6-py3-none-any.whl.metadata (5.3 kB)
Collecting torchvision (from -r requirements.txt (line 3))
Downloading torchvision-0.17.2-cp312-cp312-macosx_11_0_arm64.whl.metadata (6.6 kB)
Collecting einops (from -r requirements.txt (line 4))
Downloading einops-0.7.0-py3-none-any.whl.metadata (13 kB)
Collecting transformers>=4.25.1 (from -r requirements.txt (line 5))
Downloading transformers-4.39.2-py3-none-any.whl.metadata (134 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 134.8/134.8 kB 2.1 MB/s eta 0:00:00
Collecting safetensors>=0.3.0 (from -r requirements.txt (line 6))
Downloading safetensors-0.4.2-cp312-cp312-macosx_11_0_arm64.whl.metadata (3.8 kB)
Collecting aiohttp (from -r requirements.txt (line 7))
Downloading aiohttp-3.9.3-cp312-cp312-macosx_11_0_arm64.whl.metadata (7.4 kB)
Collecting pyyaml (from -r requirements.txt (line 8))
Downloading PyYAML-6.0.1-cp312-cp312-macosx_11_0_arm64.whl.metadata (2.1 kB)
Collecting Pillow (from -r requirements.txt (line 9))
Downloading pillow-10.2.0-cp312-cp312-macosx_11_0_arm64.whl.metadata (9.7 kB)
Collecting scipy (from -r requirements.txt (line 10))
Downloading scipy-1.12.0-cp312-cp312-macosx_12_0_arm64.whl.metadata (217 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 217.9/217.9 kB 2.3 MB/s eta 0:00:00
Collecting tqdm (from -r requirements.txt (line 11))
Downloading tqdm-4.66.2-py3-none-any.whl.metadata (57 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 57.6/57.6 kB 3.9 MB/s eta 0:00:00
Collecting psutil (from -r requirements.txt (line 12))
Downloading psutil-5.9.8-cp38-abi3-macosx_11_0_arm64.whl.metadata (21 kB)
Collecting kornia>=0.7.1 (from -r requirements.txt (line 13))
Downloading kornia-0.7.2-py2.py3-none-any.whl.metadata (12 kB)
Collecting filelock (from torch->-r requirements.txt (line 1))
Downloading filelock-3.13.3-py3-none-any.whl.metadata (2.8 kB)
Collecting typing-extensions>=4.8.0 (from torch->-r requirements.txt (line 1))
Downloading typing_extensions-4.10.0-py3-none-any.whl.metadata (3.0 kB)
Collecting sympy (from torch->-r requirements.txt (line 1))
Downloading sympy-1.12-py3-none-any.whl.metadata (12 kB)
Collecting networkx (from torch->-r requirements.txt (line 1))
Downloading networkx-3.2.1-py3-none-any.whl.metadata (5.2 kB)
Collecting jinja2 (from torch->-r requirements.txt (line 1))
Downloading Jinja2-3.1.3-py3-none-any.whl.metadata (3.3 kB)
Collecting fsspec (from torch->-r requirements.txt (line 1))
Downloading fsspec-2024.3.1-py3-none-any.whl.metadata (6.8 kB)
Collecting numpy>=1.19 (from torchsde->-r requirements.txt (line 2))
Downloading numpy-1.26.4-cp312-cp312-macosx_11_0_arm64.whl.metadata (61 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 61.1/61.1 kB 2.9 MB/s eta 0:00:00
Collecting trampoline>=0.1.2 (from torchsde->-r requirements.txt (line 2))
Downloading trampoline-0.1.2-py3-none-any.whl.metadata (10 kB)
Collecting huggingface-hub<1.0,>=0.19.3 (from transformers>=4.25.1->-r requirements.txt (line 5))
Downloading huggingface_hub-0.22.2-py3-none-any.whl.metadata (12 kB)
Collecting packaging>=20.0 (from transformers>=4.25.1->-r requirements.txt (line 5))
Downloading packaging-24.0-py3-none-any.whl.metadata (3.2 kB)
Collecting regex!=2019.12.17 (from transformers>=4.25.1->-r requirements.txt (line 5))
Downloading regex-2023.12.25-cp312-cp312-macosx_11_0_arm64.whl.metadata (40 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 40.9/40.9 kB 2.6 MB/s eta 0:00:00
Collecting requests (from transformers>=4.25.1->-r requirements.txt (line 5))
Downloading requests-2.31.0-py3-none-any.whl.metadata (4.6 kB)
Collecting tokenizers<0.19,>=0.14 (from transformers>=4.25.1->-r requirements.txt (line 5))
Downloading tokenizers-0.15.2-cp312-cp312-macosx_11_0_arm64.whl.metadata (6.7 kB)
Collecting aiosignal>=1.1.2 (from aiohttp->-r requirements.txt (line 7))
Downloading aiosignal-1.3.1-py3-none-any.whl.metadata (4.0 kB)
Collecting attrs>=17.3.0 (from aiohttp->-r requirements.txt (line 7))
Downloading attrs-23.2.0-py3-none-any.whl.metadata (9.5 kB)
Collecting frozenlist>=1.1.1 (from aiohttp->-r requirements.txt (line 7))
Downloading frozenlist-1.4.1-cp312-cp312-macosx_11_0_arm64.whl.metadata (12 kB)
Collecting multidict<7.0,>=4.5 (from aiohttp->-r requirements.txt (line 7))
Downloading multidict-6.0.5-cp312-cp312-macosx_11_0_arm64.whl.metadata (4.2 kB)
Collecting yarl<2.0,>=1.0 (from aiohttp->-r requirements.txt (line 7))
Downloading yarl-1.9.4-cp312-cp312-macosx_11_0_arm64.whl.metadata (31 kB)
Collecting kornia-rs>=0.1.0 (from kornia>=0.7.1->-r requirements.txt (line 13))
Downloading kornia_rs-0.1.2-cp312-cp312-macosx_11_0_arm64.whl.metadata (8.3 kB)
Collecting idna>=2.0 (from yarl<2.0,>=1.0->aiohttp->-r requirements.txt (line 7))
Downloading idna-3.6-py3-none-any.whl.metadata (9.9 kB)
Collecting MarkupSafe>=2.0 (from jinja2->torch->-r requirements.txt (line 1))
Downloading MarkupSafe-2.1.5-cp312-cp312-macosx_10_9_universal2.whl.metadata (3.0 kB)
Collecting charset-normalizer<4,>=2 (from requests->transformers>=4.25.1->-r requirements.txt (line 5))
Downloading charset_normalizer-3.3.2-cp312-cp312-macosx_11_0_arm64.whl.metadata (33 kB)
Collecting urllib3<3,>=1.21.1 (from requests->transformers>=4.25.1->-r requirements.txt (line 5))
Downloading urllib3-2.2.1-py3-none-any.whl.metadata (6.4 kB)
Collecting certifi>=2017.4.17 (from requests->transformers>=4.25.1->-r requirements.txt (line 5))
Downloading certifi-2024.2.2-py3-none-any.whl.metadata (2.2 kB)
Collecting mpmath>=0.19 (from sympy->torch->-r requirements.txt (line 1))
Downloading mpmath-1.3.0-py3-none-any.whl.metadata (8.6 kB)
Downloading torch-2.2.2-cp312-none-macosx_11_0_arm64.whl (59.7 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 59.7/59.7 MB 18.8 MB/s eta 0:00:00
Downloading torchsde-0.2.6-py3-none-any.whl (61 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 61.2/61.2 kB 282.2 MB/s eta 0:00:00
Downloading torchvision-0.17.2-cp312-cp312-macosx_11_0_arm64.whl (1.6 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.6/1.6 MB 20.5 MB/s eta 0:00:00
Downloading einops-0.7.0-py3-none-any.whl (44 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 44.6/44.6 kB 7.6 MB/s eta 0:00:00
Downloading transformers-4.39.2-py3-none-any.whl (8.8 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 8.8/8.8 MB 22.1 MB/s eta 0:00:00
Downloading safetensors-0.4.2-cp312-cp312-macosx_11_0_arm64.whl (393 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 393.3/393.3 kB 22.5 MB/s eta 0:00:00
Downloading aiohttp-3.9.3-cp312-cp312-macosx_11_0_arm64.whl (389 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 389.7/389.7 kB 29.7 MB/s eta 0:00:00
Downloading PyYAML-6.0.1-cp312-cp312-macosx_11_0_arm64.whl (165 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 165.6/165.6 kB 50.8 MB/s eta 0:00:00
Downloading pillow-10.2.0-cp312-cp312-macosx_11_0_arm64.whl (3.3 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 3.3/3.3 MB 24.9 MB/s eta 0:00:00
Downloading scipy-1.12.0-cp312-cp312-macosx_12_0_arm64.whl (31.4 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 31.4/31.4 MB 18.0 MB/s eta 0:00:00
Downloading tqdm-4.66.2-py3-none-any.whl (78 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 78.3/78.3 kB 60.2 MB/s eta 0:00:00
Downloading psutil-5.9.8-cp38-abi3-macosx_11_0_arm64.whl (249 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 249.9/249.9 kB 27.7 MB/s eta 0:00:00
Downloading kornia-0.7.2-py2.py3-none-any.whl (825 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 825.4/825.4 kB 18.4 MB/s eta 0:00:00
Downloading aiosignal-1.3.1-py3-none-any.whl (7.6 kB)
Downloading attrs-23.2.0-py3-none-any.whl (60 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 60.8/60.8 kB 29.4 MB/s eta 0:00:00
Downloading frozenlist-1.4.1-cp312-cp312-macosx_11_0_arm64.whl (51 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 51.9/51.9 kB 160.7 MB/s eta 0:00:00
Downloading huggingface_hub-0.22.2-py3-none-any.whl (388 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 388.9/388.9 kB 20.1 MB/s eta 0:00:00
Downloading fsspec-2024.3.1-py3-none-any.whl (171 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 172.0/172.0 kB 33.5 MB/s eta 0:00:00
Downloading kornia_rs-0.1.2-cp312-cp312-macosx_11_0_arm64.whl (1.3 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.3/1.3 MB 23.0 MB/s eta 0:00:00
Downloading multidict-6.0.5-cp312-cp312-macosx_11_0_arm64.whl (29 kB)
Downloading numpy-1.26.4-cp312-cp312-macosx_11_0_arm64.whl (13.7 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 13.7/13.7 MB 21.3 MB/s eta 0:00:00
Downloading packaging-24.0-py3-none-any.whl (53 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 53.5/53.5 kB 326.2 MB/s eta 0:00:00
Downloading regex-2023.12.25-cp312-cp312-macosx_11_0_arm64.whl (292 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 292.2/292.2 kB 20.8 MB/s eta 0:00:00
Downloading tokenizers-0.15.2-cp312-cp312-macosx_11_0_arm64.whl (2.4 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 2.4/2.4 MB 25.9 MB/s eta 0:00:00
Downloading trampoline-0.1.2-py3-none-any.whl (5.2 kB)
Downloading typing_extensions-4.10.0-py3-none-any.whl (33 kB)
Downloading yarl-1.9.4-cp312-cp312-macosx_11_0_arm64.whl (79 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 79.4/79.4 kB 82.9 MB/s eta 0:00:00
Downloading filelock-3.13.3-py3-none-any.whl (11 kB)
Downloading Jinja2-3.1.3-py3-none-any.whl (133 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 133.2/133.2 kB 22.4 MB/s eta 0:00:00
Downloading networkx-3.2.1-py3-none-any.whl (1.6 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.6/1.6 MB 22.3 MB/s eta 0:00:00
Downloading requests-2.31.0-py3-none-any.whl (62 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 62.6/62.6 kB 9.1 MB/s eta 0:00:00
Downloading sympy-1.12-py3-none-any.whl (5.7 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 5.7/5.7 MB 19.0 MB/s eta 0:00:00
Downloading certifi-2024.2.2-py3-none-any.whl (163 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 163.8/163.8 kB 16.6 MB/s eta 0:00:00
Downloading charset_normalizer-3.3.2-cp312-cp312-macosx_11_0_arm64.whl (119 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 119.4/119.4 kB 14.9 MB/s eta 0:00:00
Downloading idna-3.6-py3-none-any.whl (61 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 61.6/61.6 kB 9.3 MB/s eta 0:00:00
Downloading MarkupSafe-2.1.5-cp312-cp312-macosx_10_9_universal2.whl (18 kB)
Downloading mpmath-1.3.0-py3-none-any.whl (536 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 536.2/536.2 kB 22.7 MB/s eta 0:00:00
Downloading urllib3-2.2.1-py3-none-any.whl (121 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 121.1/121.1 kB 44.4 MB/s eta 0:00:00
Installing collected packages: trampoline, mpmath, urllib3, typing-extensions, tqdm, sympy, safetensors, regex, pyyaml, psutil, Pillow, packaging, numpy, networkx, multidict, MarkupSafe, kornia-rs, idna, fsspec, frozenlist, filelock, einops, charset-normalizer, certifi, attrs, yarl, scipy, requests, jinja2, aiosignal, torch, huggingface-hub, aiohttp, torchvision, torchsde, tokenizers, kornia, transformers
Successfully installed MarkupSafe-2.1.5 Pillow-10.2.0 aiohttp-3.9.3 aiosignal-1.3.1 attrs-23.2.0 certifi-2024.2.2 charset-normalizer-3.3.2 einops-0.7.0 filelock-3.13.3 frozenlist-1.4.1 fsspec-2024.3.1 huggingface-hub-0.22.2 idna-3.6 jinja2-3.1.3 kornia-0.7.2 kornia-rs-0.1.2 mpmath-1.3.0 multidict-6.0.5 networkx-3.2.1 numpy-1.26.4 packaging-24.0 psutil-5.9.8 pyyaml-6.0.1 regex-2023.12.25 requests-2.31.0 safetensors-0.4.2 scipy-1.12.0 sympy-1.12 tokenizers-0.15.2 torch-2.2.2 torchsde-0.2.6 torchvision-0.17.2 tqdm-4.66.2 trampoline-0.1.2 transformers-4.39.2 typing-extensions-4.10.0 urllib3-2.2.1 yarl-1.9.4
~/Projects/ai/ComfyUI master* 30s
venv ❯ venv/bin/python main.py
Total VRAM 65536 MB, total RAM 65536 MB
Set vram state to: SHARED
Device: mps
VAE dtype: torch.float32
Using sub quadratic optimization for cross attention, if you have memory or speed issues try using: --use-split-cross-attention
Import times for custom nodes:
0.0 seconds: /Users/michael/Projects/ai/ComfyUI/custom_nodes/websocket_image_save.py
Starting server
To see the GUI go to: http://127.0.0.1:8188
got prompt
model_type STABLE_CASCADE
clip missing: ['clip_g.logit_scale']
model_type STABLE_CASCADE
Missing VAE keys ['encoder.mean', 'encoder.std']
clip missing: ['clip_g.logit_scale']
Requested to load StableCascadeClipModel
Loading 1 new model
Requested to load StableCascade_C
Loading 1 new model
100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 20/20 [00:24<00:00, 1.22s/it]
Requested to load StableCascade_B
Loading 1 new model
100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 10/10 [00:20<00:00, 2.06s/it]
Requested to load StageA
Loading 1 new model
Prompt executed in 62.44 seconds
i've just tried the -pre version on a "clean" mac. having exactly the same issue. Can someone from the team reproduce it? :)
if you try: --force-fp32 does that fix it?
Does it happen to normal SD1.5 or SDXL workflows? or just Stable Cascade?
aha, interesting, before i read your posts i've crawled already present issues. And this seems to be related to a MacOS update to 14.4(.1): https://github.com/comfyanonymous/ComfyUI/issues/2992#issuecomment-2024893209
torch==2.1.2 torchvision==0.16.2
this worked for me:asdf plugin-add python
asdf install python 3.11.8
asdf local python 3.11.8
# close tab and repoen
python3 --version
python3 -m venv venv
source venv/bin/activate
pip install torch==2.1.2 torchvision==0.16.2
pip install -r requirements.txt
python main.py
but i don't know if this breaks MPS accellaration.
Does it happen to normal SD1.5 or SDXL workflows? or just Stable Cascade?
This happed to SDXL and SD 1 through 2.1.
Does it happen to normal SD1.5 or SDXL workflows? or just Stable Cascade?
This happed to SDXL and SD 1 through 2.1.
see if this helps: https://youtu.be/3RfjAIm1_Oo
Does it happen to normal SD1.5 or SDXL workflows? or just Stable Cascade?
This happed to SDXL and SD 1 through 2.1.
Oh just saw your earlier reply, looks like you already figured it out. Although I still recommend using a launcher, rather than managing python packages by hand.
- torch==2.1.2 torchvision==0.16.2
thanks,this worked for me too
- downgrade to Python 3.11
- install
torch==2.1.2 torchvision==0.16.2
this worked for me:python3 --version python3 -m venv venv source venv/bin/activate pip install torch==2.1.2 torchvision==0.16.2 pip install -r requirements.txt python main.py
Downgrading torch and torchvision worked for me as well. I was on torch=2.2.1 torchvision=0.17.1 prior, and that worked fine for SDXL and SD1.5.
Im pretty sure i've used the right models, from https://huggingface.co/stabilityai/stable-cascade/tree/main/comfyui_checkpoints
I've simply run the (untouched) workflow from the Image and that's what i'm getting:
BTW I' using a M3 Max Mac, on which I installed Comfyui per this readme including apples pytorch "tutoriual". I'm also starting the server with
--force-fp16
as per instruction.Maybe someone can nugde me in the right direction? :) I've prevously used InvokeAI wich I havent got these issues 🤔
PS: A simmilar Issue occurs when I try to create an SD-XL image (but here the output just looks like a "cloudy sky"?, but maybe thats another issue or i'ts resolved when I resolve this Issues problem)