Some first tests with making a flux kontext lora for a PS1 style. I used face to many to make a dataset with 12 image pairs. Example: https://
replicate.com/p/z2e7ern65hrm
c0cqqd4vmrkqjg
… I used the fal trainer at 1000 steps ($2.50), took about 20 mins. Prompt:
"make it a ps1 video game" Left: Pro
Right:
@fofrai
-
Fine-tuning Flux LoRA for PS1 video game style
By
–
-
SDXL PS1 LoRA enables face-to-many generation applications
By
–
face-to-many uses @artificialguybr’s original SDXL ps1 lora pic.twitter.com/Vz7987vu5V
— fofr (@fofrAI) 29 juin 2025face-to-many uses @artificialguybr
’s original SDXL ps1 lora -
Training Flux LoRA for PS1 Video Game Style
By
–
Some first tests with making a flux kontext lora for a PS1 style. I used face to many to make a dataset with 12 image pairs. Example: https://
replicate.com/p/z2e7ern65hrm
c0cqqd4vmrkqjg
… I used the fal trainer at 1000 steps ($2.50), took about 20 mins. Prompt:
"make it a ps1 video game" Left: Pro
Right: -
How Many Training Pairs Used for Model Development
By
–
How many training pairs have you been using?
-
Space Curvature Technique Enhances AI Model Prompting
By
–
"space curvature" also has a nice effect in prompts pic.twitter.com/B5G0ssaarH
— fofr (@fofrAI) 28 juin 2025"space curvature" also has a nice effect in prompts
-
AI Video Models Struggle With Realistic Tennis Generation
By
–
Wimbledon starts on Monday. Here's some terrible AI tennis from the latest video models.
— fofr (@fofrAI) 28 juin 2025
Only Hailuo 2 comes close, guy actually hits a ball convincingly. Wrong way though.
> tv footage of a men's singles tennis match at wimbledon
1. Seedream 1 pro
2. Veo 3
3. Hailuo 2
4.… pic.twitter.com/srfgIlQ3g6Wimbledon starts on Monday. Here's some terrible AI tennis from the latest video models. Only Hailuo 2 comes close, guy actually hits a ball convincingly. Wrong way though. > tv footage of a men's singles tennis match at wimbledon 1. Seedream 1 pro
2. Veo 3
3. Hailuo 2
4. -
Seedream generates realistic faces from creative prompts
By
–
faces from seedream with prompts like: > 100 yearbook photos
> 100 photos from a party photo booth
> 100 mugshots -
Evolution of AI Image Generation Models Over Time
By
–
When I was growing up, those new AI image models looked way better than the ones before it.
-
Testing Image Reference Conditions in AI Systems
By
–
I'm passing one image ref condition directly into another and seeing what happens
-
Anime Street Scene Generation with AI Prompt
By
–
With a prompt "put her on a street, make it anime"