Good question, I haven't tried larger resolutions with kontext-dev, I don't know what happens.
@fofrai
-
FLUX Model Improves Facial and Hair Texture Generation
By
–
Seems to fix FLUX chin. Adding randomness to skin texture and hair strands.
-
New Kontext LoRA for Realistic AI-Generated People
By
–
I'm working on a new kontext lora for making AI people look more like real people. These are the results of the second pass. Still a few tweaks and retraining needed, but it's getting there.
-
Veo 3 Impresses with Movie Generation Capabilities
By
–
Works well with movies too, I thought this would be impossible but I love how Veo 3 interpreted it. https://t.co/nZIg2OdP79 pic.twitter.com/ybOuYX5Yaq
— fofr (@fofrAI) 19 juillet 2025Works well with movies too, I thought this would be impossible but I love how Veo 3 interpreted it.
-
AI-Generated Video Game Character Costumes and Set Design
By
–
The costumes. The sets. The acting. 👌
— fofr (@fofrAI) 19 juillet 2025
kirby, portal, mario, pacman, sonic, minecraft https://t.co/SgSwaeDh7L pic.twitter.com/YRYiX8fF6LThe costumes. The sets. The acting. kirby, portal, mario, pacman, sonic, minecraft
-
Pushing AI Into Unconventional Territories for Discovery
By
–
You've got to push AI into the weird places, because that's where the interesting stuff is
-
Replicate seeks early API access from xAI
By
–
Replicate would love to get early access to an API Who @xai is best to talk to about this?
-
Runway Gen-4 Best Model for Identity Preservation Video
By
–
A few folks have asked me which model is best at preserving identity. If I had to pick one, purely on character consistency, I'd go with Runway's Gen-4. https://
replicate.com/runwayml/gen4-
image
… Close up outputs are accurate and you can prompt many different scenes. Unlike kontext-pro you don't -
Training and Running Fast Flux Models with LoRAs
By
–
You can train and then run from here https://
replicate.com/replicate/fast
-flux-kontext-trainer/train
… Or do you mean loras you have already trained? -
Fine-tuned LoRA Model Transforms Faces Into Character Style
By
–
I trained a kontext lora on my 0_1 webp character. The aim was to see if I could make it change any person in an input image into 0_1. https://
replicate.com/fofr/kontext-0
_1-webp
… It kind of works. It's very hit or miss. Forcing an aspect ratio change makes it more likely to work.