Qwen Image 2.0 and Qwen Image 2.0 Pro are now live on NanoGPT
We just added Qwen Image 2.0 and Qwen Image 2.0 Pro to NanoGPT.
Both are available on the Media page and support two key workflows in one model flow:
- Text-to-image generation from prompts
- Image editing by uploading an image and prompting changes
What stands out in Qwen Image 2.0
Based on Qwen's launch materials, the big improvements are:
- Stronger prompt adherence for complex instructions
- Better text and typography rendering for posters, slides, infographics, and comic-style layouts
- Native high-resolution outputs (up to 2K) for detailed scenes
- Unified generation + editing in the same model family
This is exactly the kind of model we wanted to add: one that can handle both clean prompt-following visuals and design-heavy outputs where text quality matters.
Official examples and benchmarks
If you want to inspect official visuals and benchmark charts from the Qwen launch:
- Hero/sample image: https://yqintl.alicdn.com/2893b60c7304fe38b7082af8f9b89b1ac9d14447.png
- AI Arena text-to-image benchmark chart: https://yqintl.alicdn.com/352a5d98584a6a03f6f1087854cbffae723a37e0.png
- AI Arena image-edit benchmark chart: https://yqintl.alicdn.com/a80b8e2aac2ad00d36ac87a085ff557a7163fbc3.png
Main references:
- Qwen launch post: https://qwen.ai/blog?id=qwen-image-2.0
- FAL overview page: https://fal.ai/qwen-image-2.0
Try it now
If you are doing posters, ad creatives, product mockups, or any image that depends on readable embedded text, this is one to test.
