#Rectified Flow

Discover a simplified implementation of scalable rectified flow models leveraging SD3 training techniques and LLaMA-DiT architecture. Ideal for beginners, this repository provides straightforward instructions for training on datasets such as MNIST and CIFAR, with further options for advanced exploration using ImageNet. Engage with rectified flow transformers easily, without requiring extensive prior knowledge, in a project that emphasizes accessibility and innovation in flow model methodologies.

Discover Rectified Flow: a novel approach for transport map learning and efficient data generation. It uniquely enhances Stable Diffusion with improved diversity and FID scores. Applicable in generative modeling and domain transfer, explore its capabilities with practical Colab tutorials across datasets like CIFAR10 and CelebA-HQ.

InstaFlow is an efficient one-step image generation model offering quality similar to Stable Diffusion with lower computational needs. It uses the Rectified Flow method for swift results, generating images in 0.1 seconds. Compatible with multiple pre-trained models like LoRAs and ControlNets, InstaFlow supports varied creative applications. Discover its performance on the MS COCO 2014 dataset and enjoy reduced processing time.