Qwen Releases Open-Source Instruction-Based Image Editor
The new model from Alibaba's Qwen team allows users to modify images using natural language prompts instead of complex tools or masks.
Alibaba's Qwen team has expanded its open-source portfolio with the release of Qwen-Image-Edit, a model designed for instruction-based image editing. Made available under a permissive Apache 2.0 license, the model allows users to alter specific parts of an image by providing simple text commands. The model card and weights can be found on its Hugging Face repository.
Unlike text-to-image generation models that create visuals from scratch, Qwen-Image-Edit takes an existing image and a set of instructions as input. This approach enables more controlled and targeted modifications, such as changing an object's color, adding an element to a scene, or altering a background, without requiring manual selection tools or complex software.
This release provides developers with a powerful open-source foundation for building more intuitive creative tools. By translating natural language directly into pixel-level changes, such models lower the barrier for advanced photo editing. This capability could be integrated into consumer applications, professional design software, or specialized commercial platforms.
Qwen-Image-Edit joins a growing family of multimodal models from the Qwen team, underscoring their continued investment in the open-source AI ecosystem. By providing powerful and accessible tools for both text and vision, they are helping to democratize capabilities that were once exclusive to proprietary systems.
Sources
- Visit
Qwen/Qwen-Image-Edit-2509
Hugging Face
More in Image Editing

Microsoft's Mage-Flow packs image editing into 4B
A compact model handles both text-to-image generation and instruction-based edits at native resolution, under a permissive MIT license.
Boogu-Image-0.1 Brings Unified Multimodal to Open Source
A new Apache-licensed model family folds bilingual text-to-image generation and instruction editing into one system.

SenseTime's SenseNova-Vision-7B-MoT Goes Any-to-Any
A single 7B model from SenseTime folds vision-language understanding, image generation, editing, and perception into one system.
0 comments
No comments yet. Be the first to weigh in.