Benchmarks
MMLU-ProNot evaluated
GPQA Diamond
82.9%
HLE
6.1%
LiveCodeBenchNot evaluated
SciCode
29.1%
TerminalBench Hard
28.8%
MATH-500Not evaluated
AIMENot evaluated
AIME 2025Not evaluated
IFBench
42.0%
Long Context Recall
55.3%
Tau2
99.1%
Market AverageTop Score