List your company

Geometric Reasoning on GeoAux-Bench-Core Curriculum Junior

84.75Accuracy (%)Gemini-2.5-Pro
3.827624.836345.84566.8537Mar 19, 2026

Evaluation results

MethodLinks
Gemini-2.5-ProThink=true2026.0384.75
GPT-5Think=true2026.0384.02
Gemini-2.5-FlashThink=true2026.0381.68
Qwen3-VL-235B-Ins.Think=false2026.0377.8
Qwen3-VL-30B-Ins.Think=false2026.0370.38
Doubao-seed-1.6Think=true2026.0361.42
Claude-Opus-4.1Think=true2026.0352.7
Claude-Sonnet-4.5Think=true2026.0352
Qwen3-VL-235B-Thk.Think=true2026.0350.36
Qwen3-VL-30B-Thk.Think=true2026.0349.64
MiMo-VL-7B-SFTThink=false2026.0342.13
MiMo-VL-7B-RLThink=true2026.0341.65
InternVL3-8BThink=false2026.0330.02
Qwen2.5-VL-Ins.Think=false2026.0328.81
InternVL3.5-8BThink=false2026.0326.63
GLM-4.5VThink=false2026.0324.62
GPT-4oThink=false2026.0324.29
GLM-4.1V-Thk.Think=true2026.0323.57
MathCanvas-7BThink=false2026.0318.89
BAGEL-7B-MoTThink=false2026.037.34
BAGEL-Zebra-CoTThink=false2026.036.94