Model Stability and Performance Evaluation on Benchmark dataset n=80 benchmark leaderboard · SOTA2 Research

Model Stability and Performance Evaluation on Benchmark dataset n=80

0.9895UGrok-3

Evaluation Results

Method	Links
Grok-32026.04		0.9895	0.012	0.7968	0.9069	1.8518	0.9775	0.983	0.0055
GPT-4o2026.04		0.9845	0.044	0.9597	0.9482	1.9539	0.9406	0.962	0.0215
DeepSeek-V32026.04		0.9695	0.0517	0.8594	0.953	1.9062	0.9178	0.9424	0.0246
Gemini-1.52026.04		0.9545	0.148	0.8981	0.799	1.8485	0.8065	0.8744	0.0679