Interleaved Geometry Reasoning on Interleaved Geometry Reasoning (test)
78.24AccuracyGemini-2.5-Pro
Evaluation results
| Method | Links | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Gemini-2.5-Pro2026.03 | 78.24 | 20.2 | 38.62 | 54.7 | 4.82 | 21.67 | 85.15 | 15.9 | 22.51 | 15.45 | 19.86 | |
| Faire2026.03 | 74.82 | 25.06 | 46.75 | 55.57 | 14.79 | 65.1 | 52.76 | 37.27 | 60.39 | 38.44 | 45.37 | |
| GPT-5.22026.03 | 68.78 | 2.32 | 9.77 | 44.32 | 9.85 | 43.79 | 69.23 | 7.02 | 36.59 | 30.24 | 24.62 | |
| GLM-4.1V-9B2026.03 | 63.41 | 5.86 | 15.7 | 23.97 | 0.67 | 2.98 | 98.05 | 7.43 | 22.39 | 7.04 | 12.29 | |
| Qwen3-VL-8B2026.03 | 59.71 | 16.72 | 36.71 | 54.96 | 6.32 | 27.02 | 81.5 | 8.62 | 25.47 | 10.47 | 14.85 | |
| GPT-5.12026.03 | 56.58 | 11.85 | 32.78 | 47.08 | 7.39 | 32.5 | 76.5 | 10.34 | 25.85 | 18.24 | 18.14 | |
| InternVL3.5-8B2026.03 | 45.56 | 8.57 | 27.49 | 32.81 | 4.77 | 21.33 | 85.76 | 5 | 17.84 | 5.2 | 9.35 | |
| Qwen2.5-VL-7B2026.03 | 26.34 | 3.97 | 11.6 | 16.66 | 0.6 | 2.61 | 98.27 | 5.16 | 21.83 | 3.31 | 10.1 | |
| GPT-4o2026.03 | 25.56 | 9.14 | 27.13 | 36.01 | 2.16 | 9.28 | 93.5 | 4.62 | 8.92 | 5.27 | 6.27 | |
| Kimi-VL-A3B2026.03 | 23.22 | 5 | 15.21 | 27.18 | 1.45 | 6.53 | 95.7 | 2.61 | 11.94 | 6.82 | 7.12 | |
| Gemma3-12B2026.03 | 20.97 | 7.43 | 25.35 | 33.57 | 0.83 | 3.6 | 97.56 | 1.97 | 4.13 | 0.88 | 2.33 |