Image Generation on DPG-Bench 130
88.3ScoreQwen-Image
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen-ImageModel Category=Generation-Experts2026.03 | 88.3 | |
| Seedream 3.0Model Category=Generation-Experts2026.03 | 88.3 | |
| Dynin-OmniModel Category=Unified, Native speech input/output capability (†)=true2026.03 | 86.3 | |
| Show-o2Model Category=Unified, Video generation support (*)=true2026.03 | 86.1 | |
| Lumina-DiMOOModel Category=Unified2026.03 | 86 | |
| HiDream-I1-FullModel Category=Generation-Experts2026.03 | 85.9 | |
| GPT-Image 1 [High]Model Category=Generation-Experts2026.03 | 85.2 | |
| BAGLEModel Category=Unified2026.03 | 85.1 | |
| Janus-ProModel Category=Unified2026.03 | 84.2 | |
| NExT-OMNIModel Category=Unified, Video generation support (*)=true, Native speech input/output capability (†)=true2026.03 | 84.2 | |
| FLUX.1 [Dev]Model Category=Generation-Experts2026.03 | 83.8 | |
| FudokiModel Category=Unified2026.03 | 83.6 | |
| Emu3Model Category=Unified2026.03 | 81.6 | |
| MMaDAModel Category=Unified2026.03 | 70 | |
| HyperCLOVAX-8B-OmniModel Category=Unified, Native speech input/output capability (†)=true, Reproduced results (‡)=true2026.03 | 60.8 |