Narrative Product Grid Collage Generation on Product Grid Generation 400 prompts (test)
65.115ArtiMuse ScoreSelf-Reasoning Agentic Framework
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Self-Reasoning Agentic FrameworkBackbone Model=Nano Banana, Evaluation Mode=Creation Mode, Framework=Agentic Pipeline2026.04 | 65.115 | 7.485 | 7.179 | 8.033 | |
| Self-Reasoning Agentic FrameworkBackbone Model=Seedream 4.5, Evaluation Mode=Creation Mode, Framework=Agentic Pipeline2026.04 | 64.666 | 7.234 | 7.006 | 7.747 | |
| Self-Reasoning Agentic FrameworkBackbone Model=Flux Kontext Max, Evaluation Mode=Creation Mode, Framework=Agentic Pipeline2026.04 | 64.118 | 7.157 | 7.023 | 7.908 | |
| Self-Reasoning Agentic FrameworkBackbone Model=Nano Banana Pro, Evaluation Mode=Creation Mode, Framework=Agentic Pipeline2026.04 | 63.464 | 7.512 | 7.237 | 8.077 | |
| Nano BananaBackbone Model=Nano Banana, Evaluation Mode=Creation Mode, Framework=Direct Generation2026.04 | 62.732 | 7.066 | 6.75 | 7.822 | |
| Nano Banana ProBackbone Model=Nano Banana Pro, Evaluation Mode=Creation Mode, Framework=Direct Generation2026.04 | 62.64 | 7.186 | 7.156 | 8.003 | |
| Seedream 4.5Backbone Model=Seedream 4.5, Evaluation Mode=Creation Mode, Framework=Direct Generation2026.04 | 62.367 | 6.86 | 6.65 | 7.532 | |
| Flux Kontext MaxBackbone Model=Flux Kontext Max, Evaluation Mode=Creation Mode, Framework=Direct Generation2026.04 | 56.384 | 6.444 | 4.777 | 7.544 |