Unifically LogoUnificAlly
Model logo

Qwen Image API

Text to ImageImage to Image

Alibaba image generation with multiple tiers. Max for photorealism, Plus for artistic styles, and base for general use.

Click or drag & dropPNG, JPG, WEBP, GIF · Max 100MB
Smart Prompt Rewriting
Smart prompt rewriting
Output

Your generated image will appear here

Features

What Qwen Image API offers

Text-to-image via an 800 character prompt budget (same limit as the playground)
Optional image_urls list with up to 10 files (100 MB each) for edit-style requests
Five aspect ratios: 1:1, 16:9, 9:16, 4:3, and 3:4
Negative prompts up to 500 characters
prompt_extend toggle for Alibaba smart prompt rewriting (defaults on)
Optional seed for reproducibility
Billing key alibaba/qwen-image

Use cases

Built for

Primary

Marketing - Poster and banner explorations with readable in-image typography

#2

E-commerce - Regenerate catalog frames from existing uploads when sku art shifts

#3

Editorial - Rapid ratio swaps between landscape and portrait crops

#4

Product design - Visualize CMF directions before CAD investment

#5

Localization - Iterate copy inside prompts while keeping seeds or references stable

#6

Game UI - Icon and ability art drafts grounded with negative prompts for stray props

FAQ

About Qwen Image API

It is Alibaba's base Qwen Image endpoint: you send prompt plus optional aspect_ratio, optional negative_prompt, optional image_urls for edits, optional prompt_extend, and optional seed. Pure text requests omit image_urls.

You can include up to ten image URLs per call when you need edit workflows or fused compositions.

prompt_extend forwards your prompt through Alibaba's rewriter when enabled. Disable it if you need verbatim wording preserved.

Usage maps to the alibaba/qwen-image key published on the Unifically pricing page.