Flux-V1.1-Ultra-Redux

FLUX is an open-source text-to-image generation model launched by the Black Forest Labs team, built upon an innovative multimodal architecture and a parallel diffusion Transformer structure. With 12 billion parameters (12B), this model is one of the largest open-source models of its kind.

FLUX.1 improves traditional diffusion models through flow matching methods, introducing rotational position embeddings and parallel attention layers, significantly enhancing its ability to capture image details. Within the product line, the FLUX.1 [pro] version delivers state-of-the-art image generation performance, featuring superior prompt adherence, visual quality, and output diversity.

As the latest breakthrough from the founding team of Stable Diffusion, FLUX.1 can simultaneously process multiple input types. In July of this year, it introduced the enhanced FLUX.1 Kontext version, providing high-performance, high-quality image generation solutions for creative design, content creation, and other fields.

302.AI Release date: 2024-08-02