Qwen-Image: Alibaba Foundation T2I Model with Strong Text Rendering
ComfyUI Wiki
Qwen-Image is Alibaba's foundation image generation model with strong text rendering and editing capabilities. Available in multiple quantization formats for ComfyUI.
Q
Qwen-Image
Text-to-ImageText RenderingImage EditingQuantizedFoundation text-to-image generation model from Alibaba's Qwen series. Built on a diffusion transformer architecture, it excels at complex text rendering (especially Chinese), precise image editing, and general image generation with support for diverse artistic styles. Available in BF16, FP8, FP8 mixed, NVFP4, and 2.5K resolution variants.
| Developer | Alibaba Cloud (Qwen Team) |
| Release Date | 2025-08-04 |
| Architecture | Diffusion Transformer (DiT) |
| License | Apache-2.0 |
| Text Encoder | Qwen2.5-VL-7B |
| Pipeline | Diffusers (text-to-image) |
| ControlNets | InstantX Union/Inpainting, DiffSynth Canny/Depth/Inpaint |
Guides and workflows related to this model series.
No articles found.
Comments
Sign in with GitHub to join the discussion.