Qwen-Image-2.1 arrives with better text rendering, portraits, and multi-view editing

Alibaba_Qwen · x · 2026-09-20

Alibaba's Qwen team officially announced Qwen-Image-2.1, claiming clear visual quality gains over its predecessor, especially in the aesthetics of text rendering and portraits.

The new model also balances performance across a variety of editing tasks — for example, given a three-view character reference, it can generate a complete storyboard. The post includes multiple text-rendering and editing examples.

Related event: Alibaba Open-Sources Qwen-Image-2.1, a Unified 7B Image Generation and Editing Model(20 posts)→

Original post →

More from Multimodal

Multimodal channel →