-
Welcome to the Fakes / AI / Deepfakes category
When posting AI generated content please include prompts used where possible.
dead again?
š Download finished.
/content/sdw
ā ļø Null model will be loaded, if you don't understand please uncheck the model
ERROR: ld.so: object 'libtcmalloc.so' from LD_PRELOAD cannot be preloaded (cannot open shared object file): ignored.
ERROR: ld.so: object 'libtcmalloc.so' from LD_PRELOAD cannot be preloaded (cannot open shared object file): ignored.
--2023-05-17 15:26:45--
Resolving huggingface.co (huggingface.co)... 13.35.7.102, 13.35.7.28, 13.35.7.88, ...
Connecting to huggingface.co (huggingface.co)|13.35.7.102|:443... connected.
HTTP request sent, awaiting response... 302 Found
Location: [following]
--2023-05-17 15:26:57--
Resolving cdn-lfs.huggingface.co (cdn-lfs.huggingface.co)... 13.35.24.76, 13.35.24.128, 13.35.24.38, ...
Connecting to cdn-lfs.huggingface.co (cdn-lfs.huggingface.co)|13.35.24.76|:443... connected.
HTTP request sent, awaiting response... 416 Requested Range Not Satisfiable
The file is already fully retrieved; nothing to do.
ERROR: ld.so: object 'libtcmalloc.so' from LD_PRELOAD cannot be preloaded (cannot open shared object file): ignored.
ERROR: ld.so: object 'libtcmalloc.so' from LD_PRELOAD cannot be preloaded (cannot open shared object file): ignored.
ERROR: ld.so: object 'libtcmalloc.so' from LD_PRELOAD cannot be preloaded (cannot open shared object file): ignored.
ERROR: ld.so: object 'libtcmalloc.so' from LD_PRELOAD cannot be preloaded (cannot open shared object file): ignored.
python3: can't open file '/content/sdw/launch.py': [Errno 2] No such file or directory
ā You've been running this colab for 00:05:03
š Download finished.
/content/sdw
ā ļø Null model will be loaded, if you don't understand please uncheck the model
ERROR: ld.so: object 'libtcmalloc.so' from LD_PRELOAD cannot be preloaded (cannot open shared object file): ignored.
ERROR: ld.so: object 'libtcmalloc.so' from LD_PRELOAD cannot be preloaded (cannot open shared object file): ignored.
--2023-05-17 15:26:45--
Resolving huggingface.co (huggingface.co)... 13.35.7.102, 13.35.7.28, 13.35.7.88, ...
Connecting to huggingface.co (huggingface.co)|13.35.7.102|:443... connected.
HTTP request sent, awaiting response... 302 Found
Location: [following]
--2023-05-17 15:26:57--
Resolving cdn-lfs.huggingface.co (cdn-lfs.huggingface.co)... 13.35.24.76, 13.35.24.128, 13.35.24.38, ...
Connecting to cdn-lfs.huggingface.co (cdn-lfs.huggingface.co)|13.35.24.76|:443... connected.
HTTP request sent, awaiting response... 416 Requested Range Not Satisfiable
The file is already fully retrieved; nothing to do.
ERROR: ld.so: object 'libtcmalloc.so' from LD_PRELOAD cannot be preloaded (cannot open shared object file): ignored.
ERROR: ld.so: object 'libtcmalloc.so' from LD_PRELOAD cannot be preloaded (cannot open shared object file): ignored.
ERROR: ld.so: object 'libtcmalloc.so' from LD_PRELOAD cannot be preloaded (cannot open shared object file): ignored.
ERROR: ld.so: object 'libtcmalloc.so' from LD_PRELOAD cannot be preloaded (cannot open shared object file): ignored.
python3: can't open file '/content/sdw/launch.py': [Errno 2] No such file or directory
ā You've been running this colab for 00:05:03
Started up fine for me, although I can't download the URPM model right now because Civitai is down.
Hmm I tried adding the code,it's giving me an error
Cannot save changes
ā¤ļøāš„ oversimplified fork of NoCrypt's colab
Visitors
Important website status:
gradio civitai hf
Start š
[5]
Install Configurations ?
latest_webui:
Configurations
output_to_drive:
configs_in_drive:
fast_start:
merge_in_vram:
colab_optimizations:
ram_patch_for_sd2:
dpmpp_v2:
verbose_download:
commandline_arguments:
--enable-insecure-extension-access --opt-sdp-attention --share --no-half-vae --disable-safe-unpickle --no-hashing --gradio-queue --xformers
commit_hash:
Insert text here
ngrok_token:
Insert text here
ngrok_region:
jp
Models & VAEs
original_1_5:
sd_vae:
controlnet:
none
null_model:
warning: ngrok token not detected
š Download finished.
/content/sdw
Python 3.10.11 (main, Apr 5 2023, 14:15:10) [GCC 9.4.0]
Version: v1.2.1
Commit hash: 89f9faa63388756314e8a1d96cf86bf5e0663045
Installing requirements
Launching Web UI with arguments: --enable-insecure-extension-access --opt-sdp-attention --share --no-half-vae --disable-safe-unpickle --no-hashing --gradio-queue --xformers
2023-05-17 17:08:08.818593: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2023-05-17 17:08:10.822423: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
reading checkpoint metadata: /content/sdw/models/Stable-diffusion/ChilloutMix.safetensors: AssertionError
Traceback (most recent call last):
File "/content/sdw/modules/sd_models.py", line 62, in __init__
self.metadata = read_metadata_from_safetensors(filename)
File "/content/sdw/modules/sd_models.py", line 232, in read_metadata_from_safetensors
assert metadata_len > 2 and json_start in (b'{"', b"{'"), f"{filename} is not a safetensors file"
AssertionError: /content/sdw/models/Stable-diffusion/ChilloutMix.safetensors is not a safetensors file
Checkpoint anything-v4.5-pruned.safetensors not found; loading fallback ChilloutMix.safetensors
Loading weights [None] from /content/sdw/models/Stable-diffusion/ChilloutMix.safetensors
loading stable diffusion model: SafetensorError
Traceback (most recent call last):
File "/content/sdw/modules/sd_models.py", line 419, in get_sd_model
load_model()
File "/content/sdw/modules/sd_models.py", line 452, in load_model
state_dict = get_checkpoint_state_dict(checkpoint_info, timer)
File "/content/sdw/modules/sd_models.py", line 273, in get_checkpoint_state_dict
res = read_state_dict(checkpoint_info.filename)
File "/content/sdw/modules/sd_models.py", line 252, in read_state_dict
pl_sd = safetensors.torch.load_file(checkpoint_file, device=device)
File "/usr/local/lib/python3.10/dist-packages/safetensors/torch.py", line 259, in load_file
with safe_open(filename, framework="pt", device=device) as f:
safetensors_rust.SafetensorError: Error while deserializing header: HeaderTooLarge
Not sure how to fix this. Any suggestions?
elif select_model == "ChilloutMix":
print("Downloading ChilloutMix about 4GB")
gdown.download(url="https://civitai.com/api/download/models/6424?type=Model&format=SafeTensor&size=full&fp=fp16", output=f"{modelpath}/ChilloutMix.safetensors", quiet=False, fuzzy=True)
gdown.download(url="https://civitai.com/api/download/models/6424?type=Config&format=Other", output=f"{modelpath}/ChilloutMix.yaml", quiet=False, fuzzy=True)
This is what I added
There's mostly only 2 causes for errors. Either civitai is down, which is quite often or you have caused some error/typo.
Code looks good tho. Got the same error a couple of times aswell, and I don't know exactly what causes it because it seems kinda random. Sometimes switching/loading models works and sometimes not. You can try the pruned version, this worked for me at the first time I got this error.
Had exact stuff today aswell but I simply completely reloaded the collab runtime and it worked again.
Currently I'm looking for a solution to host models on drive again and load them into the collab since I like changing models quite often but honestly it's on a low priority right now since I mostly run a1111 locally. Sometimes it's quite nice having multiple webuis running.
can i use NoCrypt remastered colab instead? I've tried running it and also put your urpm code, seems working but I haven't actually tested/generate anything from it
Yeah its fine, as long as you know what you are doing, the colab i provided is a simplified version of that
When downloading large model in colab from civitai sometimes it randomly drop the connection which causes the downloaded model to be corrupted, it will appear as the download have been completed.
The url you added is not a "Chilloutmix" model, which causes the error, the url refer to a hypernetwork model called "mechanicus"
here are steps to use the facialized Lora for anyone that wants to make cumshot fakes:
1. follow all the steps in the tutorial and boot up the website
2. now go onto img2img Inpainting and remember to select the Inpainting model
3. delete the default prompts for nudifying
4. paste these prompts below:
positive prompt: cum, sticky white cum, large cum amounts, facial, brown hair girl gets facialized, facial
negative prompt: lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, anime, drawing, illustration
5. tap the red button underneath "generate" then select Lora and scroll and select facialized
6. upload your image and mask the face
7. make sure all the settings below are done:
Mask: Inpaint Masked
Masked Content: original
Inpaint At Full Resolution: Checked (make sure this check box is checked) (if u can't find this then don't worry because i didnt find it)
select Crop and Resize: Also a very important value to set
Sampling Steps: 55 is a good value, but you can play around with it (I personally use 100)
Sampling Method: Euler A (Works great, but feel free to play around and experiment with it)
Width: 768
Height: 768
restore faces: off
deionizing strength: this is how much the picture is changed and I personally use 0.05 but you can play around with it (0.05 - 0.2 is a good starting point) (higher=more change, lower=less change)
CFG Scale: This is how much cum there is, I use around 12.5 - 20 but you can play around with it (higher=more cum, lower=less cum)
8. generate your image and enjoy!
1. follow all the steps in the tutorial and boot up the website
2. now go onto img2img Inpainting and remember to select the Inpainting model
3. delete the default prompts for nudifying
4. paste these prompts below:
positive prompt: cum, sticky white cum, large cum amounts, facial, brown hair girl gets facialized, facial
negative prompt: lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, anime, drawing, illustration
5. tap the red button underneath "generate" then select Lora and scroll and select facialized
6. upload your image and mask the face
7. make sure all the settings below are done:
Mask: Inpaint Masked
Masked Content: original
Inpaint At Full Resolution: Checked (make sure this check box is checked) (if u can't find this then don't worry because i didnt find it)
select Crop and Resize: Also a very important value to set
Sampling Steps: 55 is a good value, but you can play around with it (I personally use 100)
Sampling Method: Euler A (Works great, but feel free to play around and experiment with it)
Width: 768
Height: 768
restore faces: off
deionizing strength: this is how much the picture is changed and I personally use 0.05 but you can play around with it (0.05 - 0.2 is a good starting point) (higher=more change, lower=less change)
CFG Scale: This is how much cum there is, I use around 12.5 - 20 but you can play around with it (higher=more cum, lower=less cum)
8. generate your image and enjoy!
Last edited:
If you download the lora from the code i provided Don't forget to add <lora:superfacialized:1> to the prompt
Hmm OK cause this is the link I was trying to add..
having trouble finding this? 5. tap the red button underneath "generate" then select Lora and scroll and select facialized
Try selecting entire face.
Because you have Masked Content set to Original the full image is processed along with the mask meaning the face should still come out looking similar to the original just with changes from the network