FireFlow: Fast Inversion of Rectified Flow for Image Semantic Editing

Yingying Deng, Xiangyu He, Changwang Mei, Peisong Wang, Fan Tang
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:13110-13128, 2025.

Abstract

Though Rectified Flows (ReFlows) with distillation offer a promising way for fast sampling, its fast inversion transforms images back to structured noise for recovery and following editing remains unsolved. This paper introduces FireFlow, an embarrassingly simple yet effective zero-shot approach that inherits the startling capacity of ReFlow-based models (such as FLUX) in generation while extending its capabilities to accurate inversion and editing in 8 steps. We first demonstrate that a carefully designed numerical solver is pivotal for ReFlow inversion, enabling accurate inversion and reconstruction with the precision of a second-order solver while maintaining the practical efficiency of a first-order Euler method. This solver achieves a $3\times$ runtime speedup compared to state-of-the-art ReFlow inversion and editing techniques while delivering smaller reconstruction errors and superior editing results in a training-free mode. The code is available at this-URL.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-deng25c, title = {{F}ire{F}low: Fast Inversion of Rectified Flow for Image Semantic Editing}, author = {Deng, Yingying and He, Xiangyu and Mei, Changwang and Wang, Peisong and Tang, Fan}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {13110--13128}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/deng25c/deng25c.pdf}, url = {https://proceedings.mlr.press/v267/deng25c.html}, abstract = {Though Rectified Flows (ReFlows) with distillation offer a promising way for fast sampling, its fast inversion transforms images back to structured noise for recovery and following editing remains unsolved. This paper introduces FireFlow, an embarrassingly simple yet effective zero-shot approach that inherits the startling capacity of ReFlow-based models (such as FLUX) in generation while extending its capabilities to accurate inversion and editing in 8 steps. We first demonstrate that a carefully designed numerical solver is pivotal for ReFlow inversion, enabling accurate inversion and reconstruction with the precision of a second-order solver while maintaining the practical efficiency of a first-order Euler method. This solver achieves a $3\times$ runtime speedup compared to state-of-the-art ReFlow inversion and editing techniques while delivering smaller reconstruction errors and superior editing results in a training-free mode. The code is available at this-URL.} }
Endnote
%0 Conference Paper %T FireFlow: Fast Inversion of Rectified Flow for Image Semantic Editing %A Yingying Deng %A Xiangyu He %A Changwang Mei %A Peisong Wang %A Fan Tang %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-deng25c %I PMLR %P 13110--13128 %U https://proceedings.mlr.press/v267/deng25c.html %V 267 %X Though Rectified Flows (ReFlows) with distillation offer a promising way for fast sampling, its fast inversion transforms images back to structured noise for recovery and following editing remains unsolved. This paper introduces FireFlow, an embarrassingly simple yet effective zero-shot approach that inherits the startling capacity of ReFlow-based models (such as FLUX) in generation while extending its capabilities to accurate inversion and editing in 8 steps. We first demonstrate that a carefully designed numerical solver is pivotal for ReFlow inversion, enabling accurate inversion and reconstruction with the precision of a second-order solver while maintaining the practical efficiency of a first-order Euler method. This solver achieves a $3\times$ runtime speedup compared to state-of-the-art ReFlow inversion and editing techniques while delivering smaller reconstruction errors and superior editing results in a training-free mode. The code is available at this-URL.
APA
Deng, Y., He, X., Mei, C., Wang, P. & Tang, F.. (2025). FireFlow: Fast Inversion of Rectified Flow for Image Semantic Editing. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:13110-13128 Available from https://proceedings.mlr.press/v267/deng25c.html.

Related Material