Qwen: Qwen Image 3 Pro

Qwen Image 3 Pro is an image generation and editing model from Qwen, listed on api14 under the model id qwen/qwen-image-3-pro. It sits in the image category, and its provider describes it as supporting precise rendering of text and details as small as 10px, along with richer world knowledge. If you are building a pipeline that needs to produce or modify images from a prompt, this is the model family to evaluate.

The model accepts text and image inputs and returns images. Its context window is 65,536 tokens, with a maximum output of 58,982 tokens. Image output is controlled through a set of image parameters: resolution is an enum of 1K or 2K, aspect ratio is an enum covering 1:1, 1:2, 1:4, 2:1, 2:3, 3:2, 3:4, 4:1, 4:3, 4:5, 5:4, 9:16 and 16:9, and n is a range from 1 to 6 images per request. You can also pass up to 4 input references, which is what makes editing and reference-guided generation possible, and seed is available as a boolean.

Those capabilities point to a few practical patterns. Text-to-image generation with a chosen resolution and aspect ratio covers straightforward asset creation. Because the model accepts image inputs alongside text, it also fits editing and reference-driven work, where you supply one or more input references and describe the change you want. The aspect ratio enum matters for teams producing assets for fixed layouts, such as square social images, vertical story formats or wide banners, and the n parameter lets you request several variations in one call rather than issuing separate requests. The emphasis on small text rendering suggests use cases where legible lettering inside the image is the point, such as posters, labels or UI mockups.

You call it through api14's OpenAI-compatible API using the model id qwen/qwen-image-3-pro. The supported request parameters are max_tokens, presence_penalty, response_format, seed, temperature and top_p, alongside the image parameters described above (resolution, aspect_ratio, n and input_references). If your existing code already targets an OpenAI-compatible endpoint, switching to this model is largely a matter of changing the model id and adding the image-specific fields to the request body.

Keep the boundaries in mind when choosing. This is an image model: it outputs images, not text, so tasks that need a written answer belong with a text model. The context window is 65,536 tokens, and input references are capped at 4, with n capped at 6 images per request. If your workflow needs more references or a larger batch in a single call, plan to split the work across multiple requests.