Agentic Image Generation on IA-Bench 1.0 (test)
73.3Checklist Accuracy (Plan)SCOPE
Evaluation results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| SCOPEModel Category=Agentic Image Generation Model, MLLM Backbone=GPT-5.5-0424, Image Generation and Edit Backbone=Qwen-Image-2.0, Feedback loop attempts=up to 32026.06 | 73.3 | 45.2 | 44.4 | 45.2 | 46.7 | 30 | 23.3 | 9 | 30.9 | |
| Qwen-Image-AgentModel Category=Agentic Image Generation Model, MLLM Backbone=GPT-5.5-0424, Image Generation and Edit Backbone=Qwen-Image-2.0, Feedback loop attempts=up to 32026.06 | 72.9 | 65.5 | 67.6 | 73.6 | 45.3 | 43.7 | 46.1 | 49 | 45.4 | |
| Seedream-5.0-LiteModel Category=Closed-source Image Generation Model2026.06 | 71.3 | 58.3 | 50.1 | 66.4 | 46 | 37 | 21.1 | 48 | 36 | |
| GEMSModel Category=Agentic Image Generation Model, MLLM Backbone=GPT-5.5-0424, Image Generation and Edit Backbone=Qwen-Image-2.0, Feedback loop attempts=up to 32026.06 | 70.6 | 28.4 | 49.4 | 52.6 | 41.3 | 18.3 | 18.9 | 13 | 24.9 | |
| Nano BananaModel Category=Closed-source Image Generation Model2026.06 | 68 | 63.9 | 61.7 | 60.2 | 42 | 43.3 | 42.2 | 48 | 43.1 | |
| Nano Banana ProModel Category=Closed-source Image Generation Model2026.06 | 60.8 | 66.2 | 68.3 | 72 | 32.7 | 44.3 | 47.8 | 52 | 42.6 | |
| MindBrushModel Category=Agentic Image Generation Model, MLLM Backbone=GPT-5.5-0424, Image Generation and Edit Backbone=Qwen-Image-2.0, Feedback loop attempts=up to 32026.06 | 56.1 | 51.8 | 53.6 | 53.1 | 28 | 32.7 | 35.6 | 13 | 30.2 | |
| GPT-Image-1.5Model Category=Closed-source Image Generation Model2026.06 | 55.1 | 55.6 | 55.2 | 87.6 | 23.3 | 36.7 | 35 | 72 | 35.7 | |
| Qwen-Image-2.0Model Category=Closed-source Image Generation Model2026.06 | 50 | 48.2 | 38 | 51.8 | 20 | 27.7 | 6.7 | 11 | 17.4 | |
| GenSearcherModel Category=Agentic Image Generation Model, MLLM Backbone=GPT-5.5-0424, Image Generation and Edit Backbone=Qwen-Image-2.0, Feedback loop attempts=up to 32026.06 | 37 | 30.1 | 46.5 | 46.6 | 9.3 | 20.3 | 24.4 | 11 | 17.3 | |
| Qwen-ImageModel Category=Open-source Image Generation Model2026.06 | 30 | 28.2 | 35.1 | 41.1 | 4.7 | 17.7 | 6.1 | 9 | 9.4 | |
| FLUX.2-devModel Category=Open-source Image Generation Model2026.06 | 29.4 | 23.2 | 33.1 | 52.9 | 5.3 | 15 | 9.4 | 11 | 10 | |
| Bagel w/ CoTModel Category=Open-source Image Generation Model, Chain of Thought=true2026.06 | 22.6 | 26.7 | 12.8 | 5.9 | 2 | 19 | 0.6 | 0 | 6.5 | |
| Echo-4oModel Category=Open-source Image Generation Model, Chain of Thought=false2026.06 | 22.1 | 11.6 | 17.1 | 7.9 | 0.7 | 4 | 0.6 | 0 | 1.6 | |
| BagelModel Category=Open-source Image Generation Model, Chain of Thought=false2026.06 | 20 | 12.6 | 15.1 | 4.7 | 0.7 | 4 | 0.6 | 0 | 1.6 | |
| SD-3.5-largeModel Category=Open-source Image Generation Model2026.06 | 19 | 9.2 | 24.2 | 10.5 | 0 | 5.7 | 6.1 | 1 | 3.6 | |
| Echo-4o w/ CoTModel Category=Open-source Image Generation Model, Chain of Thought=true2026.06 | 17.4 | 10.1 | 9.3 | 7.6 | 0 | 4 | 0.6 | 0 | 1.4 | |
| SD-3.5-mediumModel Category=Open-source Image Generation Model2026.06 | 15.6 | 6.9 | 20.4 | 5.9 | 0 | 4 | 3.3 | 0 | 2.2 |