Creator
OpenAI
Released
2025-12-11
Intelligence
46.6
Artificial Analysis Index
Coding
44.2
Artificial Analysis Index
In $/1M
$1.75
input tokens
Out $/1M
$14.00
output tokens
Blended $/1M
$4.81
3:1 blended
Speed
0
tokens / sec
Benchmark breakdown
Independent evaluation scores, normalized to 0–100.
GPQA Diamond
86
Humanity’s Last Exam
25
MMLU-Pro
86
SciCode
46
LiveCodeBench
89
MATH-500—
AIME 2025
97
τ²-Bench (agentic)
74
Terminal-Bench Hard
43
IFBench
65