Not for commercial use.

Model republished from: https://huggingface.co/shuttleai/shuttle-3-diffusion-fp8

Shuttle 3 Diffusion is a text-to-image AI model designed to generate rich and detailed images from text cues in just 4 steps. It delivers improved performance in image quality, typography, complex cues understanding, and resource efficiency.

Training details

Shuttle 3 Diffusion uses Flux.1 Schnell as its foundation. It can produce images similar to Flux Dev or Pro in just 4 steps and is licensed under Apache 2. The model was partially refined during training. When used beyond 10 steps, it enters a “refinement mode,” enhancing image details without changing the composition. We have overcome the limitations of the Schnell series models by using a special learning method, resulting in improved detail and colors.