Qwen: Meet Qwen-Image-2.1, now open weights

Launch post: 7B unified generation/editing model that Qwen says 'outperforms most closed-source models'; native RGBA; up to 10 reference images; links to blog, GitHub, ModelScope and Hugging Face.

Metrics read 2026-09-30 via api.fxtwitter.com: 2,271,861 views.

Meet Qwen-Image-2.1, the most balanced and cost-effective image generation model in the Qwen-Image series! Now open weights! 🎨

A unified model for both generation and editing, delivering top-tier quality in a lightweight package.

Highlights: 👀

• Compact & exceptionally fast: A lightweight 7B architecture that outperforms most closed-source models, with drastically accelerated inference for multi-image inputs.
• Native transparency: Natively generates and edits RGBA layers, enabling seamless compositing and text editing within transparent images.
• Versatile, high-fidelity editing: Supports up to 10 reference images and precise local control while preserving strict fidelity for portraits and products.
• Broad coverage & stunning aesthetics: Excels at panoramas, infographics, and virtual try-ons, delivering realistic textures and elegant typography.

Start to create your next masterpiece with Qwen-Image-2.1! 🖼️

• Blog: https://qwen.ai/blog?id=qwen-image-2.1
• GitHub: https://github.com/QwenLM/Qwen-Image-2.1
• Model Scope: https://www.modelscope.cn/models/Qwen/Qwen-Image-2.1
• Hugging Face: https://huggingface.co/Qwen/Qwen-Image-2.1

Media: https://pbs.twimg.com/media/HSqWS--aYAAi5G0.jpg?name=orig

views 2271893 · likes 8260 · reposts 887 · replies 294 (at fetch time)

Archived 2026-09-29 via fxtwitter (unofficial).