Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
22.9Peak Accuracy
3
Apr 7, 2026
82.6Normalized Log Accuracy
3
Mar 18, 2026
72.1Normalized Log Accuracy
3
Mar 18, 2026
63.5Normalized Log Accuracy
3
Mar 18, 2026
87.5Normalized Log Accuracy
3
Mar 18, 2026
0.96F1 Score
3
Feb 26, 2026
48.25Accuracy
3
Feb 26, 2026
2,245.5TPS
3
Feb 26, 2026
1,986.3TPS
3
Feb 26, 2026
2,768.5TPS
3
Feb 26, 2026
2,979.8Throughput (TPS)
3
Feb 26, 2026
75.69Accuracy
3
Feb 26, 2026
78.7Accuracy (Zebralogic)
3
Feb 26, 2026
83.3Pass@1
3
Feb 26, 2026
0.04Total Cost (USD)
3
Feb 26, 2026
0.11Total Cost (USD)
3
Feb 26, 2026
0.09Total Cost (USD)
3
Feb 26, 2026
0.06Total Cost (USD)
3
Feb 26, 2026
3ER
3
Feb 26, 2026
66ACC1
3
Feb 26, 2026
80.6Score
2
Jul 7, 2026
10.649Acc (Prediction D'_pre)
2
Jul 2, 2026
3.116Accuracy (Prediction of D'_pre)
2
Jul 2, 2026
9.886Accuracy (Prediction D'_pre)
2
Jul 2, 2026
98.5Quality Score
2
Jun 26, 2026