GitHub radar
Qwen Image 2.1: generate and edit images in one model
Qwen Image 2.1 is a new open model from Alibaba that generates images from text descriptions and edits existing ones in the same pipeline.
Qwen-Image-2.1 is a new model from Alibaba's Qwen team that handles both text-to-image generation and image editing in one pipeline. You give it a text description and it creates an image, or you give it an existing image and a written instruction and it changes a specific area. For editing, you mark the area with a brush or a mask and describe what to put there. If you need to keep a person's face or a specific product consistent, you provide up to 10 reference photos and the model preserves the identity. The model also generates images with transparent backgrounds, which is useful for content layering and product cutouts. It supports resolutions up to 2752x1536. It runs locally via ComfyUI - Comfy-Org has published a ready workflow for this model.
Why a vibe-coder should care
For routine content work - posts, backgrounds, product images - the quality is enough, especially if you do not have a subscription to a paid service. The ability to edit a specific area without regenerating the whole image is a strong side of this model. There is a quality gap compared to paid services, but for everyday content tasks this is not significant.
How to install
Copy this and send it to your agent — Claude Code, Codex, any of them:
Install Qwen Image 2.1 locally via ComfyUI: https://huggingface.co/Comfy-Org/Qwen-Image-2.1 - set it up through the ComfyUI model manager and show me how to generate an image from a text prompt.
A regular laptop with 16 GB of RAM or more - the quantized version takes about 4 GB of memory.
Open on Hugging Face▌ More finds