AI RESEARCH
Qwen-Image-VAE-2.0 Technical Report
arXiv CS.CV
•
ArXi:2605.13565v1 Announce Type: new 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