Black Forest Labs, a newly formed US-based AI startup with German roots, has unveiled its first suite of text-to-image models, FLUX.1.
Founded by key developers behind Stable Diffusion, the company aims to push the boundaries of image and video generation.
The launch comes shortly after Stability AI’s controversial release of Stable Diffusion 3, which faced criticism for its poor human anatomy generation. Notably, three former Stability AI engineers—instrumental in developing Stable Diffusion—are among the founders of Black Forest Labs.
FLUX.1 offers three models: a high-end commercial option, a mid-range model with open weights for non-commercial use, and a faster open-source version. Black Forest Labs claims superior image quality and better adherence to text prompts compared to competitors like Midjourney and DALL-E.
Key Features:
Cutting-edge output quality, second only to our state-of-the-art model
FLUX.1 [pro].Competitive prompt following, matching the performance of closed source alternatives .
Trained using guidance distillation, making
FLUX.1 [dev]more efficient.Open weights to drive new scientific research, and empower artists to develop innovative workflows.
Generated outputs can be used for personal, scientific, and commercial purposes as described in the flux-1-dev-non-commercial-license.
Try it: https://replicate.com/black-forest-labs/flux-pro
Ref: https://huggingface.co/black-forest-labs/FLUX.1-dev
Git: https://github.com/black-forest-labs/flux
Investors: Marc Andreessen and Ben Horowitz https://a16z.com/announcement/investing-in-black-forest-labs/
