gm
a new exploration from Frags, a branch of Project Native (which explores how AI image-to-video systems behave when left to their own devices)
I set particular visual conditions, then watch as the model, unprompted, exposes its perceptual grammar — gestures like Optics Logic, Phantom 3D, and Fabricated Interpolation are shown here
the imagery draws from both documentary and New Topographics traditions — from everyday infrastructure to sites shaped by conflict, collapse, or neglect (and with figures uncommonly included this time) — not as spectacle but as terrain where the machine articulates its own sense of reality, and may serve as a mirror for us
Whoa, I just tested the IC Edit @huggingface demo and it seems the new 🐐👑 of image editing for
It's an image editing LoRA for FLUX featuring:
👨🎤 Identity preservation (beating GPT-4o)
✏️✏️ Does multiple edits
🐎 10s image editing
🌌 style support