Plane Geometry Problem Solving on Formalgeo (test)
0.857AccuracyMLLM Interpreter
Evaluation Results
| Method | Links | |
|---|---|---|
| MLLM InterpreterTraining Data Size=5.5k2026.01 | 0.857 | |
| Gemini2.5-ProTraining Data Size=-2026.01 | 0.818 | |
| Gemini2.5-FlashTraining Data Size=-2026.01 | 0.805 | |
| DFE-GPSTraining Data Size=238k2026.01 | 0.753 | |
| GLM4.1-VTraining Data Size=-2026.01 | 0.734 | |
| Claude-Sonnet-4Training Data Size=-2026.01 | 0.691 | |
| Claude-Opus-4.1Training Data Size=-2026.01 | 0.691 | |
| GeoUniTraining Data Size=235k2026.01 | 0.598 | |
| GPT-4oTraining Data Size=-2026.01 | 0.58 | |
| Qwen2.5-VL 32BTraining Data Size=-2026.01 | 0.573 |