Qwen-Image-VAE-2.0 Technical Report

TL;DR

Qwen-Image-VAE-2.0 introduces advanced high-compression VAEs with improved architecture, training on billions of images, and a new benchmark, achieving state-of-the-art reconstruction and diffusability, especially in text-rich scenarios.

Contribution

The paper presents Qwen-Image-VAE-2.0 with novel architecture enhancements, large-scale training, and a new benchmark, advancing high-compression image reconstruction and diffusion capabilities.

Findings

Achieves state-of-the-art reconstruction performance.

Excels in text-rich scenarios with high compression.

Demonstrates superior diffusability and faster convergence.

Abstract

We present Qwen-Image-VAE-2.0, a suite of high-compression Variational Autoencoders (VAEs) that achieve significant advances in both reconstruction fidelity and diffusability. To address the reconstruction bottlenecks of high compression, we adopt an improved architecture featuring Global Skip Connections (GSC) and expanded latent channels. Moreover, we scale training to billions of images and incorporate a synthetic rendering engine to improve performance in text-rich scenarios. To tackle the convergence challenges of high-dimensional latent space, we implement an enhanced semantic alignment strategy to make the latent space highly amenable to diffusion modeling. To optimize computational efficiency, we leverage an asymmetric and attention-free encoder-decoder backbone to minimize encoding overhead. We present a comprehensive evaluation of Qwen-Image-VAE-2.0 on public reconstruction benchmarks. To evaluate performance in text-rich scenarios, we propose OmniDoc-TokenBench, a new benchmark comprising a diverse collection of real-world documents coupled with specialized OCR-based evaluation metrics. Qwen-Image-VAE-2.0 achieves state-of-the-art reconstruction performance, demonstrating exceptional capabilities in both general domains and text-rich scenarios at high compression ratio. Furthermore, downstream DiT experiments reveal our models possess superior diffusability, significantly accelerating convergence compared to existing high-compression baselines. These establish Qwen-Image-VAE-2.0 as a leading model with high compression, superior reconstruction, and exceptional diffusability.