Creator
Alibaba
Released
2025-09-05
Intelligence
26.1
Artificial Analysis Index
Coding
25.5
Artificial Analysis Index
In $/1M
$1.20
input tokens
Out $/1M
$6.00
output tokens
Blended $/1M
$2.40
3:1 blended
Speed
57
tokens / sec
Benchmark breakdown
Independent evaluation scores, normalized to 0–100.
GPQA Diamond
76
Humanity’s Last Exam
9
MMLU-Pro
84
SciCode
37
LiveCodeBench
65
MATH-500—
AIME 2025
75
τ²-Bench (agentic)
33
Terminal-Bench Hard
20
IFBench
48