List your company

Geometric Diagram Generation on GenExam-Math (test)

8Structural CorrectnessGPT-Image-1
-0.321.8446.16Feb 3, 2026

Evaluation results

MethodLinks
GPT-Image-1Model Category=Closed-source Models2026.02852
Socratic-Generator-ImageModel Category=Open-source Unified MLLMs, Base Model=Qwen-Image2026.02642.4
Seedream 4.0Model Category=Closed-source Models2026.022.639.8
Imagen-4-UltraModel Category=Closed-source Models2026.022.635.9
Gemini-2.5-Flash-ImageModel Category=Closed-source Models2026.020.743.1
Seedream 3.0Model Category=Closed-source Models2026.020.718.6
FLUX.1 Kontext maxModel Category=Closed-source Models2026.02023.5
Qwen-ImageModel Category=Open-source T2I Models2026.02018.9
HiDream-I1-FullModel Category=Open-source T2I Models2026.02016.7
FLUX.1 devModel Category=Open-source T2I Models2026.02012.2
FLUX.1 KreaModel Category=Open-source T2I Models2026.0207
Stable Diffusion 3.5 LargeModel Category=Open-source T2I Models2026.02012.2
BAGEL (thinking)Model Category=Open-source Unified MLLMs2026.02011.7
BAGELModel Category=Open-source Unified MLLMs2026.02014.7
Show-o2-7BModel Category=Open-source Unified MLLMs2026.02010.8
Show-o2-1.5B-HQModel Category=Open-source Unified MLLMs2026.0207.3
BLIP3o-NEXT-GRPO-Text-3BModel Category=Open-source Unified MLLMs2026.02015.5
BLIP3o-8BModel Category=Open-source Unified MLLMs2026.0206.4
Janus-ProModel Category=Open-source Unified MLLMs2026.02013.7
Emu3Model Category=Open-source Unified MLLMs2026.02011.3