Benchmarks
MMLU-ProNot evaluated
GPQA Diamond
84.5%
HLE
16.1%
LiveCodeBenchNot evaluated
SciCode
38.2%
TerminalBench HardNot evaluated
MATH-500Not evaluated
AIMENot evaluated
AIME 2025Not evaluated
IFBenchNot evaluated
Long Context Recall
58.7%
Tau2Not evaluated
Market AverageTop Score