Flow Map Distillation Without Data

TL;DR

This paper introduces a data-free flow map distillation method that samples solely from the prior distribution, avoiding data dependency and mismatch risks, and achieves state-of-the-art results in image generation.

Contribution

The authors propose a novel data-free framework for flow map distillation that predicts the teacher's sampling path and corrects errors, surpassing data-based methods.

Findings

Achieves FID of 1.45 on ImageNet 256x256 with 1 step

Surpasses all data-based flow distillation methods

Establishes a new state-of-the-art in flow model distillation

Abstract

State-of-the-art flow models achieve remarkable quality but require slow, iterative sampling. To accelerate this, flow maps can be distilled from pre-trained teachers, a procedure that conventionally requires sampling from an external dataset. We argue that this data-dependency introduces a fundamental risk of Teacher-Data Mismatch, as a static dataset may provide an incomplete or even misaligned representation of the teacher's full generative capabilities. This leads us to question whether this reliance on data is truly necessary for successful flow map distillation. In this work, we explore a data-free alternative that samples only from the prior distribution, a distribution the teacher is guaranteed to follow by construction, thereby circumventing the mismatch risk entirely. To demonstrate the practical viability of this philosophy, we introduce a principled framework that learns to predict the teacher's sampling path while actively correcting for its own compounding errors to ensure high fidelity. Our approach surpasses all data-based counterparts and establishes a new state-of-the-art by a significant margin. Specifically, distilling from SiT-XL/2+REPA, our method reaches an impressive FID of 1.45 on ImageNet 256x256, and 1.49 on ImageNet 512x512, both with only 1 sampling step. We hope our work establishes a more robust paradigm for accelerating generative models and motivates the broader adoption of flow map distillation without data.