Qwen Image
Neural Network
Qwen Image is an image generation model from the Qwen series, strong at rendering complex text within images.
Specifications
- Input
- text, images
- Output
- images
- Aspect ratio
- 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3
- File formats
- WEBP, JPG, PNG
How it works Qwen Image?
Qwen Image is an image generation model from the Qwen series, focused on complex scenes and significantly advanced in handling text on images: inscriptions, captions, and short lines are readable and compositionally appropriate. This expands the range of tasks—from posters and covers to layouts with text. Through BotHub, the model opens in the same window as other generators: access without VPN or foreign cards, payment with Russian cards in rubles as you go, and purchased tokens do not expire. You can switch between models directly in the interface or Telegram bot without building a new stack. Designers use Qwen Image for posters, ad creatives, and covers where readable headlines are important; marketers for quick banner and product card variations; content creators for article and presentation illustrations; product teams for mockups and promo materials. A unified OpenAI-compatible API allows integrating generation into your service alongside other BotHub models. Data is transmitted via an encrypted channel and not stored. For companies, contracts, invoices, EDI, and admin panel limits are available.Frequently asked questions about Qwen Image
You can use generated results for commercial purposes. You own all rights to the content you create. The only restriction: make sure your prompt does not include copyrighted third-party material. You are responsible for respecting the rights to any input data.
Qwen Image is an image generation model from Qwen. It creates images from text descriptions and can also work with input images: you upload a reference and get a new variation. Suitable for illustrations, concepts, covers, banners, and social media visuals.
Exact resolutions and formats depend on request parameters—they are specified in the page's technical header and API documentation. You get a finished raster image based on your description or uploaded reference. The easiest way to find the right option is by experimenting: make a few generations and compare.
Reasoning before answering is not marked in our data, which doesn't deny the possibility—it's just not noted. Qwen Image's task is different: to turn your description into an image. If you need a chain of thought and task analysis, switch to a text model in the same window.
There are no notes about function calling or structured JSON output in our data, so we cannot guarantee them. For scenarios involving tools and strict response schemas, use a text model, and leave image generation to Qwen Image—both work via the unified OpenAI-compatible BotHub API.
Images—yes: image-to-image mode is confirmed, so you can attach a reference and ask for a variation or rework. There are no notes about audio or video input in our data, so don't count on them; if you need audio or video analysis, choose a specialized model from the catalog.
The prompt language is not specified in our data, so we won't promise anything specific—check it with your task, it takes a couple of minutes. What definitely doesn't depend on the language: the BotHub interface, support, and payment with Russian cards, without VPN or foreign cards.