We introduce OmniFlow, a novel generative model designed for any-to-any generation tasks such as text-to-image, text-to-audio, and audio-to-image synthesis. OmniFlow advances the rectified flow (RF) framework used in text-to-image models to handle the joint distribution of multiple modalities. It outperforms previous any-to-any models on a wide range of tasks, such as text-to-image and text-to-audio synthesis. Our work offers three key contributions: First, we extend RF to a multi-modal setting and introduce a novel guidance mechanism, enabling users to flexibly control the alignment between different modalities in the generated outputs. Second, we propose a novel architecture that extends the text-to-image MMDiT architecture of Stable Diffusion 3 and enables audio and text generation. The extended modules can be efficiently pretrained individually and merged with the vanilla text-to-image MMDiT for fine-tuning. Lastly, we conduct a comprehensive study on the design choices of rectified flow transformers for large-scale audio and text generation, providing valuable insights into optimizing performance across diverse modalities. The Code will be available atthis https URL.
View on arXiv@article{li2025_2412.01169,
title={ OmniFlow: Any-to-Any Generation with Multi-Modal Rectified Flows },
author={ Shufan Li and Konstantinos Kallidromitis and Akash Gokul and Zichun Liao and Yusuke Kato and Kazuki Kozuka and Aditya Grover },
journal={arXiv preprint arXiv:2412.01169},
year={ 2025 }
}