-
ICMag with help from Landrace Warden and The Vault is running a NEW contest in November! You can check it here. Prizes are seeds & forum premium access. Come join in!
AI will take over Hollywood
- Thread starter Weedhopper
- Start date
Weedhopper
Well-known member
I follow a guy on FB. Robert Crowell and a few others.
Weedhopper
Well-known member
See how real she looks.
AI is changing everything.
AI is changing everything.
xtsho
Well-known member
Sucks to be a graphic designer these days as anyone can create art with a text prompt.
Weedhopper
Well-known member
I'm guessing Terminator movies are not that far fetched anymore.
xtsho
Well-known member
I'm guessing Terminator movies are not that far fetched anymore.
I think were a long ways off from that. But who knows? The Military already has a working AI piloted fighter jet.
Weedhopper
Well-known member
And Drones that can kick ass without putting a soldier in harms way.
Weedhopper
Well-known member
Weedhopper
Well-known member
xtsho
Well-known member
I don't know if I have a pro setup. I'm just running the software locally instead of using an online service. My system only has a 16gb Nvidia card so I can only do so much with it as it's GPU intensive.
Most of these images look like they're done on one of many online sites where you enter a text prompt and they generate the images for you. Those sites are using at a minimum an Nvidia A100 with 80gb of RAM. Those are $18,000 cards and they likely have multiple or they just rent processing power from large GPU data centers.
If you do a search for "text to image generator" you'll get a list of sites you can use to create images. Or if you use the Microsoft Edge browser with Co-Pilot it has an image creator built in.
My goal when I started working with AI a few years back was more for machine learning and training LLM's. The image creation is neat but I'm more focused on the backend and training my own models for specific subjects and tasks.
Depending on your system you can also run Automatic1111 which is a web interface for Stable-Diffusion you can use instead of entering prompts at a command line. It runs on Windows and Linux. It's a one click install on Windows. It will run with a 4gb GPU but it will be real slow at processing images.
I just downloaded a different 14gb model from Hugging Face. That's a large repository of open source models and datasets. The thing about running locally is that you have the ability to set parameters that are unavailable using most of the online services. You can really tweak the image effect to your liking.
The model I just downloaded is called Mobius and so far it seems to be pretty good at making a girl in a bikini on the beach. What did you want me to make?
GitHub - AUTOMATIC1111/stable-diffusion-webui: Stable Diffusion web UI
Stable Diffusion web UI. Contribute to AUTOMATIC1111/stable-diffusion-webui development by creating an account on GitHub.
github.com
Hugging Face – The AI community building the future.
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
huggingface.co
mudballs
Well-known member
L
Idk i was thinking a super happy golden retriever, smoking a blunt, like johnny depp in Leaving Las Vegas, and the wasted lawyer in the backseat?I don't know if I have a pro setup. I'm just running the software locally instead of using an online service. My system only has a 16gb Nvidia card so I can only do so much with it as it's GPU intensive.
Most of these images look like they're done on one of many online sites where you enter a text prompt and they generate the images for you. Those sites are using at a minimum an Nvidia A100 with 80gb of RAM. Those are $18,000 cards and they likely have multiple or they just rent processing power from large GPU data centers.
If you do a search for "text to image generator" you'll get a list of sites you can use to create images. Or if you use the Microsoft Edge browser with Co-Pilot it has an image creator built in.
My goal when I started working with AI a few years back was more for machine learning and training LLM's. The image creation is neat but I'm more focused on the backend and training my own models for specific subjects and tasks.
Depending on your system you can also run Automatic1111 which is a web interface for Stable-Diffusion you can use instead of entering prompts at a command line. It runs on Windows and Linux. It's a one click install on Windows. It will run with a 4gb GPU but it will be real slow at processing images.
I just downloaded a different 14gb model from Hugging Face. That's a large repository of open source models and datasets. The thing about running locally is that you have the ability to set parameters that are unavailable using most of the online services. You can really tweak the image effect to your liking.
The model I just downloaded is called Mobius and so far it seems to be pretty good at making a girl in a bikini on the beach. What did you want me to make?
View attachment 19012635
View attachment 19012632
GitHub - AUTOMATIC1111/stable-diffusion-webui: Stable Diffusion web UI
Stable Diffusion web UI. Contribute to AUTOMATIC1111/stable-diffusion-webui development by creating an account on GitHub.
Hugging Face – The AI community building the future.
We’re on a journey to advance and democratize artificial intelligence through open source and open science.huggingface.co
Dime
Well-known member
Gotta agree
Attachments
Weedhopper
Well-known member
