List your company

Geometric Reasoning on GeoAux-Bench Core (Total)

83.16AccuracyGemini-2.5-Pro
9.184828.389947.59566.8001Mar 19, 2026

Evaluation results

MethodLinks
Gemini-2.5-ProThink=true2026.0383.16
GPT-5Think=true2026.0380.62
Gemini-2.5-FlashThink=true2026.0379.53
Qwen3-VL-235B-Ins.Think=false2026.0374.85
Qwen3-VL-30B-Ins.Think=false2026.0370.25
Doubao-seed-1.6Think=true2026.0365
Qwen3-VL-235B-Thk.Think=true2026.0362.25
Qwen3-VL-30B-Thk.Think=true2026.0358.75
Claude-Opus-4.1Think=true2026.0355.82
Claude-Sonnet-4.5Think=true2026.0355.03
MiMo-VL-7B-SFTThink=false2026.0350.87
MiMo-VL-7B-RLThink=true2026.0350.7
GLM-4.5VThink=false2026.0340.24
InternVL3.5-8BThink=false2026.0336.26
InternVL3-8BThink=false2026.0335.17
Qwen2.5-VL-Ins.Think=false2026.0333.25
GLM-4.1V-Thk.Think=true2026.0331.76
GPT-4oThink=false2026.0328.13
MathCanvas-7BThink=false2026.0324.63
BAGEL-7B-MoTThink=false2026.0312.95
BAGEL-Zebra-CoTThink=false2026.0312.03