Freepik, the web graphic design platform, unveiled a brand new “open” AI picture mannequin on Tuesday that the corporate says was skilled completely on commercially licensed, “safe-for-work” photos.
The mannequin, known as F Lite, accommodates round 10 billion parameters — parameters being the inner elements that make up the mannequin. F Lite was developed in partnership with AI startup Fal.ai and skilled utilizing 64 Nvidia H100 GPUs over the course of two months, in accordance with Freepik.
F Lite joins a small and rising assortment of generative AI fashions skilled on licensed information.
We’ve been secretly engaged on this for months! It feels good to lastly share it!
LINKS:
• Regular model: extra predictable and prompt-faithful, however much less inventive:
• Texture model: is extra chaotic and error-prone, however delivers higher textures and… pic.twitter.com/GX5mIpYE8O
— Javi Lopez ⛩️ (@javilopen) April 29, 2025
Generative AI is on the heart of copyright lawsuits in opposition to AI firms, together with OpenAI and Midjourney. It’s regularly developed utilizing large quantities of content material — together with copyrighted content material — from public sources across the internet. Most firms growing these fashions argue truthful use shields their apply of utilizing copyrighted information for coaching with out compensating the homeowners. Many creators and IP rights holders disagree.
Freepik has made two flavors of F Lite obtainable, normal and texture, each of which had been skilled on an inner information set of round 80 million photos. Standard is extra predictable and “prompt-faithful,” whereas texture is extra “error-prone” however delivers higher textures and artistic compositions, in accordance with the corporate.
Here’s a picture from the usual mannequin generated with the immediate “An individual standing in entrance of a sundown, in majestic environment.”
Freepik makes no declare that F Lite produces photos superior to main picture mills like Midjourney’s V7, Black Forest Labs’ Flux household, or others. The purpose was to make a mannequin overtly obtainable in order that builders might tailor and enhance it, in accordance with the corporate.
That being stated, operating F Lite is not any simple feat. The mannequin requires a GPU with at the very least 24GB of VRAM.
Other firms growing media-generating fashions on licensed information embody Adobe, Bria, Getty Images, Moonvalley, and Shutterstock. Depending on how AI copyright lawsuits shake out, the market might develop exponentially.
