List your company

Geometry Problem Solving on PyGeoX Easy

62AccuracyQwen3-8B-RL
51.654.35759.7Jun 8, 2026

Evaluation results

MethodLinks
Qwen3-8B-RLInference=RL-trained, Evaluation Protocol=zero-shot2026.0662
Proprietary-CInference=reasoning, Evaluation Protocol=zero-shot2026.0661
Proprietary-BInference=reasoning, Evaluation Protocol=zero-shot2026.0655
Proprietary-AInference=default, Evaluation Protocol=zero-shot2026.0653
DeepSeek-V3.2Inference=default, Evaluation Protocol=zero-shot2026.0652