Example of negative prompts

• nsfw, worst quality, low quality, depth of field, blurry, greyscale, monochrome, 3D face, nose, cropped, lowres, text, jpeg artifacts, signature, watermark, username, blurry, artist name, trademark, watermark, title, tan, muscular, loli, petite, child, infant, toddlers, chibi, sd character, multiple view, Reference sheet
• NSFW, lowres, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts,signature, watermark, username, blurry, artist name, humans, castle, solo, bad face, deform face, nude, scantily clad, deform hands, nude, naked,  text, cropped, out of frame, worst quality, low quality, jpeg artifacts, ugly, duplicate, morbid, mutilated, extra fingers, mutated hands, poorly drawn hands, poorly drawn face, mutation, deformed, blurry, dehydrated, bad anatomy, bad proportions, extra limbs, cloned face, disfigured, gross proportions, malformed limbs, missing arms, missing legs, extra arms, extra legs, fused fingers, too many fingers, long neck,easynegative, ng_deepnegative_v1_75t, negative_hand, bad_pictures

Complete Comfy UI SDXL 1.0 Guide

• https://huggingface.co/spaces/SpacesExamples/ComfyUI

Reviewing Leonardo.ai

App.leonardo.ai is an innovative Image content generation process assisted by a powerful artificial intelligence (AI) offering a wide range of features and capabilities.

The interface is modern, very much inspired by discord in the Web 3.0 style

In terms of performance, app.leonardo.ai delivers impressive results. For people not able to run SD at home, Leonardo is a good way to access several good quality models such as SDXL, RPG4, RealisticVision, Dreamshaper, etc ...

To access Leonardo.Ai, you can simply sign up by visiting their landing page and clicking on the ‘Create an account‘ or ‘Start using Leonardo‘ button. 

If this is your first time signing up, enter your first name and email and click the ‘Count me in’ button to be added to the early access waiting list. 

However, if you’ve already been granted access, you can click the ‘Yes, I’m whitelisted’ button which will take you to the web-app login.

Generating images will cost tokens (and the daily amount is 150) by tweaking the settings generating can cost as low as 1 token (disabling Prompt Magic, and Alchemy) otherwise a generation can cost 16 token, on a free plan all generated images will be made public.

An access to the community pages will list all generated images and if one catch your eyes it can be used as a base for a remix, or img2img ; other options such as 'remove background' or 'upscale' are also available.

Here is an example of a generated image:

with this settings:

Full body Portrait of Beautiful toned Megan Fox, high quality illustration in post apocalyptic , muscular, messy hair, award winning photo shoot 

and model 'Dreamshaper v7'

AI Garbage generated show

 

AI is still learning how to understand and generate humor, and it often doesn't get it right. 

The jokes can be forced, unnatural, or just plain unfunny.

There are a few reasons why AI-generated comedy can be so hit-or-miss. 

First, AI doesn't have the same understanding of human humor as a  we do. 

It doesn't know what makes people laugh, and it doesn't know how to build on existing jokes or comedic tropes.

Second, AI is often trained on a very limited dataset of text and code. 

This means that it's only exposed to a narrow range of humor, and it doesn't have the ability to generate new and original jokes.

Finally, AI is still under development, and it's constantly learning and improving. This means that the quality of AI-generated comedy is constantly changing, and it's possible that it will eventually get better.

However, for now, AI-generated comedy is still a work in progress. If you're looking for a good laugh, you're probably better off watching a comedy show that was written by humans.

And wait there's more

AI has been around for about less than a year, though 

https://fablestudio.github.io/showrunner-agents/

Feels like these shows are like elevator music, endless generated content, it is like watching the 10,000 monkeys typing on their typewriters, and waiting for a good sentence.

What I got from the show : AI will please the lazy, and will take the responsibilities off us when doing mistakes, and because of the law of diminishing returns, what the AI are going to produce is always going to be of less quality compared to what we can write, imagine or produce.

Png extracting workflow for ComfyUI (Better version )

requires PIL 

 so maybe do a 

 $pip install PIL 

 beforehand

import argparse
import PIL.Image
import PIL.PngImagePlugin

def extract_text(filename):
    """Extracts text from a PNG file."""
    img = PIL.Image.open(filename)
    info = img.text["workflow"]
    return info

## usage python readpng.py ~/Downloads/example.png > comfy_workflow.json

def main():
    """Parses arguments and extracts text from a PNG file."""
    parser = argparse.ArgumentParser(description="Extracts text from a PNG file.")
    parser.add_argument("filename", help="The PNG file to extract text from.")
    args = parser.parse_args()
    text = extract_text(args.filename)
    print(text)

if __name__ == "__main__":
    main()

Extracting metadata from a png file in Python

 A very simple program using PIL to extract the png text embedded as metadata by comfyUI 

import PIL.Image
import argparse

def extract_text(filename):
    png = PIL.Image.open(filename)
    return png.text

if __name__ == "__main__":
    parser = argparse.ArgumentParser(description='png metadata infor')
    parser.add_argument('filename', help='png File to check')
    args = parser.parse_args()
    text = extract_text(args.filename)
    print(text)

Google Bard is finally there

 Here are my take on the new Google AI:

Google has recently announced a new AI language model called Bard. 

Bard is a factual language model from Google AI, trained on a massive dataset of text and code. It can generate text, translate languages, write different kinds of creative content, and answer your questions in an informative way.

Beware though as it is still under development, but it has already shown great promise. 

Here are some of the key features of Bard:

• Factual language model: Bard is trained on a massive dataset of text and code, which allows it to generate text that is both informative and accurate.
• Multilingual: Bard can translate between over 100 languages, which makes it a valuable tool for communication and collaboration.
• Creative: Bard can write different kinds of creative content, such as poems, code, scripts, musical pieces, email, letters, etc. This makes it a powerful tool for artists, writers, and developers.
• Intelligent: Bard is able to understand and respond to natural language queries. This makes it a valuable tool for research and learning.
• Answers can be listened to via an audio option
• Responses can be shared on reddit, linkedn, or exported into Google Docs

I believe that Bard has the potential to be a powerful tool for many different purposes. It could be used to:

• Improve communication and collaboration between people who speak different languages.
• Create new and innovative ways to learn and educate.
• Generate creative content that is both informative and engaging.
• Answer your questions in an informative way, even if they are open ended, challenging, or strange.

I am excited to see how Bard develops in the future. 

Happy Accidents saves the day (again)!

Playing with 'Happy Accident '

happyaccidents.ai/image

for the prompt and details (model, lora, etc ...)

Error downloading audio: get_throttling_function_name: could not find match for multiple

 In the past I had posted this https://alashazam.blogspot.com/2023/05/python-program-to-download-music.html explaining how to download video using python library pytube

If you are still using this program, you will notice that it is recently not functional anymore 

As of end of June 2023, this program will crash with this error:

Error downloading audio: get_throttling_function_name: could not find match for multiple

and if you are investigating further you will also discover this error

pytube.exceptions.RegexMatchError: get_throttling_function_name: could not find match for multiple

During handling of the above exception, another exception occurred:

To fix the error - you will have to edit the file /lib/python3.10/site-packages/pytube/cipher.py

on line +287 and remove the ; character

so the new file will be like this

Once that change is made the pytube library will work again.

Installing a local chatGPT on Linux

It all start with this link

Happy Accidents saves the day!

Since the ban of free google account using colab for generating AI artworks, 

Many users have upgraded their google account, others have left, and after using free online services, they might get frustrated to not having access to text inversion and Loras/Lycoris modules. 

There is a new deal as recently https://www.happyaccidents.ai/ is adding these modules for free as new features

The results are quite good:

see for yourself

How To Display Code In Your Blogger Website Using Highlight JS

Python program to download music

Here is a working python3 program able to download songs from youtube

source in text below

Deviantart and AI Artworks

DeviantArt is a popular online platform that allows artists and creators to share their work with a global audience. 

The website has a vast collection of digital art, traditional art, photography, and other creative works, making it an excellent resource for inspiration and networking. 

DeviantArt also has a strong community aspect, with groups and forums dedicated to various art styles, mediums, and interests. This allows users to connect with like-minded artists, get feedback on their work, and collaborate on projects. 

Additionally, DeviantArt offers various tools and resources for artists, such as tutorials, stock images, and brushes, making it a valuable resource for both beginners and professionals. 

Recently with the AI trends DeviantArt has implemented an innovative approach to AI-generated artwork. The website has a feature called "AI Generated Art" that allows users to upload artwork that has been created with the help of AI algorithms. 

This includes artwork generated with neural networks, machine learning, and other AI techniques. The artwork is tagged as "AI Generated" and is given a unique copyright status, which allows artists to retain ownership of their work while acknowledging the contribution of the AI algorithms. This approach is designed to recognize the role of AI in the creative process while ensuring that artists are properly recognized and compensated for their work. It's an interesting and forward-thinking approach that highlights the potential of AI in the creative arts.

Overall, DeviantArt is a fantastic platform for artists and anyone interested in the creative arts, and is definitely worth checking out.

But there is this little trick I found quite useful, when wanting to credit an artist for a picture, after downloading an image, you will in the file name a code that can be pasted into an url shortener and get back to the place it was taken from.

for example this file 'just_a_cute_cat_in_a_red_bag_by_trumanbrown_dfjtqpo-pre'

correspond to this picture :

Like this picture

it contains this code 'dfjtqpo'

and when adding this code to the url shortener fav.me/

like so fav.me/dfjtqpo

you will be redirected to the creator's page 

Pretty neat

How to get a prompt from a png file

It is sometime frustrating to see a beautiful AI generated pictures on reddit for example, and the user did not post the prompt or the detail on the model used.

With the help of this website (pngchunk.com)

you will be able to see the prompt used and all AI details (assuming the file has been made with Stable Diffusion)

Taking AI animation to the NEXT level

GPG error gpg: problem reading source

 GPG is a really useful tool to encrypt sensitive files, but when an error like 

error gpg: problem reading source

that means the file is corrupt, and it means unless you have backup the file is lost.

Comparison portrait combining Scarlett Johansson, Kristen Bell, Christina Aguilera, Angelina Jolie, Heather Graham

 A while ago this question was asked on this blog

What if a crazy scientist would create the perfect model by combining :

Scarlett Johansson

Kristen Bell

Christina Aguilera

Angelina Jolie

Heather Graham

Now with the assistance of AI (stable diffusion) and this website

it is possible to get several portraits:

The prompt used:

one perfect female face mixed of (Scarlett Johansson, Kristen Bell, Christina Aguilera, Angelina Jolie, Heather Graham), looking at viewer 

ControlNet

The whole playlist

makepose.com

Useful to put characters into a situation or a specific movement

https://makepose.com/ 

Using the website above, there is a 2D or 3D model

everything is controlled by the mouse, joint movement, zoom in and out (scrolling button) and rotation

it is possible to add more character, and even to generate the pose into a pseudo stable diffusion module.

• prompt: str = Form(""), num_samples: int = Form(1), image_resolution: int = Form(512),
• low_threshold: int = Form(100), high_threshold: int = Form(200), ddim_steps: int = Form(20),
• scale: float = Form(9.0), seed: Optional[int] = Form(None), eta: float = Form(0.0),
• a_prompt: str = Form("best quality, extremely detailed"), n_prompt: str = Form("longbody, lowres, bad anatomy, bad hands, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality"), detect_resolution: int = Form(512),

Other special features:

• Scrolling and zooming
• Download button 
• Dragging characters around 
• up to 4 characters
• Double clicking a limb to increase limb size

This is the perfect complement to use Controlnet via  Stable Diffusion