InstaFlow: One Step is Enough for High-Quality Diffusion-Based Text-to-Image Generation

Introducing InstaFlow: A game-changer in text-to-image generation! This one-step diffusion model, leveraging Rectified Flow's 'reflow' technique, achieves SD-level image quality in milliseconds. With an FID of 23.3 on MS COCO 2017-5k and training taking just 199 GPU days, InstaFlow sets new standards in speed and quality. Paper link: https://arxiv.org/abs/2309.06380 You can also read: https://arxiv.org/abs/2209.03003 Table of Content: 00:00 Intro 00:23 Diffusion model 05:06 Rectified Flow 08:44 Reflow 13:19 Text-Conditioned Distillation 14:51 CFG Velocity 16:31 Experiments and Results Icon made by Freepik from flaticon.com