A cinematic photo depicting a vast, snow-covered mountain range, with towering peaks piercing the clouds. The scene exudes a sense of grandeur and technological advancement, capturing the essence of the post's content.
Created using FLUX.1 with the prompt, "A cinematic photo depicting a vast, snow-covered mountain range, with towering peaks piercing the clouds. The scene exudes a sense of grandeur and technological advancement, capturing the essence of the post's content. "

Black Forest Labs Unveils FLUX.1 – A SotA Open-Weights Text-to-Image Model

Black Forest Labs has unveiled FLUX.1 [schnell], a 12 billion parameter rectified flow transformer that generates images from text descriptions. This open-source model pushes the boundaries of text-to-image synthesis with its cutting-edge output quality and competitive prompt following.

FLUX.1 leverages latent adversarial diffusion distillation to produce high-quality images in just 1 to 4 steps. Released under the Apache-2.0 license, it’s available for personal, scientific, and commercial use.

Developers can access a reference implementation and sampling code in a dedicated GitHub repository. The model is also accessible via API from bfl.ml, replicate.com, fal.ai, and through ComfyUI for local inference with a node-based workflow.

As part of the broader FLUX.1 suite, [schnell] is optimized for local development and personal use. Its siblings include FLUX.1 [pro], offering top-tier performance for professional applications, and FLUX.1 [dev], an open-weight model for non-commercial use.

The FLUX.1 family employs a hybrid architecture of multimodal and parallel diffusion transformer blocks. By incorporating flow matching, rotary positional embeddings, and parallel attention layers, these models improve upon previous diffusion approaches.

Black Forest Labs’ team, comprising distinguished AI researchers and engineers, aims to make advanced generative AI widely accessible. They believe this approach fosters innovation, increases transparency, and builds trust in the technology.

With $31 million in Series Seed funding led by Andreessen Horowitz, Black Forest Labs is poised to continue advancing generative AI. Their advisory board includes industry veterans like Michael Ovitz and Prof. Matthias Bethge, bringing extensive experience in content creation and AI research.

FLUX.1 represents a significant step forward in open-source text-to-image synthesis. As the foundation for future developments, including text-to-video systems, it showcases Black Forest Labs’ commitment to pioneering accessible, high-performance generative AI tools.

Linkedin

CONTACT