Side by side
See the difference.
Scores, capabilities and costs for the models you’re considering.
| 1 model | alibabaQwen Image 3 Text to ImageRemove × |
|---|---|
| What’s different | Accepts Chinese or English prompts plus negative prompts; optional prompt rewriting and seeded output up to 2K-square pixels. |
| Date | Jul 21, 2026Released |
API prices USD · provider configuration shown below | |
| Listed rate | $0.04/ imagefal · all routes ↗1K |
| Provider endpoints | 2 endpoints → |
Benchmarks Each result retains its evaluation configuration | |
| AA Image ↗2.1 | 1,070.63 ↗Text-to-image · all categoriesIndependent result95% CI 1,058.63–1,082.63 · n=5,848 |
Specifications | |
| Context window | 65.5K |
| Max output | 59K |
| Parameters | — |
| Structured output | Supported |
| Image input | Supported |