What is Qwen Image 2.1 Turbo?
Qwen Image 2.1 Turbo is an image generation and editing model from the Qwen team. Released on October 9, 2026, Turbo uses an eight-step denoising schedule. It can create an image from a written description or edit supplied images using a natural-language instruction. The same Qwen model can also generate PNG images with a transparent alpha channel when the prompt asks for transparency.
On this page, Qwen Image 2.1 Turbo is selected for your first visit. One generated image uses one purchased PhotoGenerAI credit. You can change the model in the tool when you want to compare an idea with GPT Image 2 or Nano Banana 2. The selected model and its credit requirement are shown before generation, so your next image follows your choice.
Write a Qwen image prompt that gives clear directions
Begin with the subject and what it is doing. Then describe the scene, composition, lighting, and visual style. For a product image, specify the material, visible details, camera angle, and background. For an illustration, define the shape, palette, and line treatment. Concrete visual instructions give Qwen Image 2.1 Turbo a clearer job than a long list of quality labels.
For an edit, separate the change from what should stay. For example, ask Qwen to replace the background while keeping the product's color, outline, and label. If you want a substantial restyle, describe the new treatment and the source details that matter. Review the result before the next edit, and change one instruction at a time when you need to understand which part of the prompt is affecting the image.
Use up to ten reference images
Qwen Image 2.1 Turbo accepts up to ten uploaded images in this tool. The first image is the starting subject for an edit; the remaining images can guide appearance, objects, clothes, or composition. Refer to them by order in your prompt. Say that image one supplies the subject, image two supplies the outfit, and image three supplies the setting when those are their intended roles.
A reference provides visual context. Describe how Qwen should use it instead of expecting the model to infer the relationship between every upload. Keep the set relevant to one image task, with a clear subject and consistent instructions. If you only want a new image from text, leave the upload area empty. You can also use our image-to-image tool when your project starts with an existing image and a transformation.
Generate transparent PNG images with Qwen
Use explicit transparency language. Qwen recommends the form: "This is an RGBA image with transparency" followed by your subject description, then "The image has alpha channel and the background is transparent." Include the framing and the details you need around the subject. Transparent backgrounds are useful for a sticker, a product asset, or an illustration that you want to place over another design.
Download the PNG to keep any alpha channel in the Qwen result. A white background in a preview is not evidence of transparency; inspect the downloaded image over a contrasting background. Transparency comes from the model's output and prompt, rather than a separate background-removal switch. If you want a subject placed in a complete scene, describe that scene instead of asking for an alpha background.
Read the official Qwen model card for the release and downloadable model. For another starting point, explore our image to image tool or AI photo editor.