Alibaba
Qwen Image 3.0
Best for multilingual text, infographics and packaging, with reference edits.
Or run a ready prompt
- A bilingual tea shop menu board in English and Chinese with six drinks and prices, chalk on slate, photo
- A clean infographic poster titled "HOW A BEE MAKES HONEY" with five numbered steps and simple icons
- Packaging for a jasmine green tea with the brand name "MOON GARDEN" on a pale green paper box, product photo

What Qwen Image 3.0 is good at
Qwen Image 3.0 is Alibaba’s image model, strongest at text in many languages and at layouts with a lot of information. Phosphene runs both versions through Alibaba’s Model Studio: Pro for the final image, and the standard model for the same generation at a lower price.
Strong at
- Text in many languages, including Chinese, inside the image
- Infographics, menus and packaging with a lot of information
- Edits guided by up to three reference images
- 1K and 2K output on both versions
Keep in mind
- Up to three reference images; choose Nano Banana 2.1 or Muse Image for more
- Ranks below GPT Image 2.5, Nano Banana 2.1 and Muse Image on blind quality votes
Tips for Qwen Image 3.0
- Write the text in the language you want it rendered in, inside quotes.
- For infographics, list the sections in order and say how many items each one has.
- Draft on the standard version, then run the prompt you like on Pro for the final.
How Qwen Image 3.0 ranks
In the Artificial Analysis Image Arena, Qwen Image 3.0 Pro ranks 17th for text-to-image and 18th for image editing; the standard Qwen Image 3.0 places 18th for text-to-image. Source: Artificial Analysis (artificialanalysis.ai), as of October 16, 2026.
Versions in Phosphene
- From 4 credits
Qwen Image 3.0 Pro
Alibaba Qwen Image 3.0 Pro via Model Studio. Multilingual text, 1K/2K, and up to three reference edits.
- From 3 credits
Qwen Image 3.0
Alibaba Qwen Image 3.0 via Model Studio. Same 1K/2K generation and reference edits as Pro at a lower price.
Qwen Image 3.0 specs
| Pro | Standard | |
|---|---|---|
| Price per image | From 4 credits | From 3 credits |
| Resolution | 1K, 1.2K, 1.5K | 1K, 1.2K, 1.5K |
| Aspect ratios | 1:1, 3:4, 4:3, 9:16, 16:9, 3:2, 2:3 | 1:1, 3:4, 4:3, 9:16, 16:9, 3:2, 2:3 |
| Reference photos | Up to 3 | Up to 3 |
| Typical time | About 12 seconds | Not measured yet |
Qwen Image 3.0 price per image
Credits for one image from a prompt, at each setting the composer offers.
Qwen Image 3.0 Pro
| Resolution | Credits |
|---|---|
| 1K | 4 |
| 1.2K / 1.5K | 7 |
Reference photos add up to 1 credit in total.
Qwen Image 3.0
| Resolution | Credits |
|---|---|
| 1K / 1.2K / 1.5K | 3 |
One reference photo does not change the price; more photos add up to 1 credit in total.
The composer shows the exact price before you generate.
Questions about Qwen Image 3.0
- How much does Qwen Image 3.0 cost in Phosphene?
- Qwen Image 3.0 Pro: 4 credits at 1K, 7 at 1.2K or 1.5K. Qwen Image 3.0: 3 credits at 1K or 1.2K or 1.5K. Qwen Image 3.0 Pro: reference photos add up to 1 credit in total. Qwen Image 3.0: one reference photo does not change the price; more photos add up to 1 credit in total. The composer shows the exact price before you generate.
- What is the difference between Qwen Image 3.0 Pro and the standard version?
- Both generate at 1K or 2K and accept up to three reference images. Pro ranks higher on blind quality votes; the standard version costs less per image, so it suits drafts.
- Can Qwen Image 3.0 write Chinese and other languages?
- Yes. Multilingual text is its strongest side. Put the exact words in quotes, in the language you want.
- Do I need an Alibaba Cloud account or API key?
- No. Phosphene runs the model for you and charges credits per image from your Phosphene balance.
Other image models
- GPT Image 2.5Prompt adherence, readable text, product and brand images
- Nano Banana 2.1Clear text, detailed images and edits with up to 14 references
- Nano Banana 2High-quality images fast, clear text and reference edits
- Seedream 5Campaign images, multilingual typography and realism
- MAI-Image 2.6Photoreal people, complex scenes and accurate text
- Muse ImageReadable text, accurate layouts and multi-reference edits at a low price
- Krea 3Fast edits that change one thing and keep the rest of the photo
- Grok ImagineTypography, posters, precise edits and polished campaign images
- Ideogram 4.5Posters, logos and exact text, with edits that keep the framing
- FLUX 3Text-rich layouts and photoreal scenes, plus Klein for fast, low-cost drafts