---
license: mit
---

# Adversarial Flow Models

This repository contains the official checkpoins of the paper [Adversarial Flow Models](https://www.arxiv.org/abs/2511.22475).

Adversarial Flow Models is a class of generative models that unifies Adversarial Models and Flow Models. This repo contains the pretrained ImageNet-256px models in the paper experiment section.

## Usage

Code and instructions are available in the [ByteDance-Seed/Adversarial-Flow-Models](https://github.com/ByteDance-Seed/Adversarial-Flow-Models) repository.

## Content 
* `models/` contains pre-trained ImageNet256px checkpoints.
* `eval/` contains pre-generated 50k samples for evaluations following ADM npz format.
* `misc/` contains VAE and other checkpoints used in training.