Creator
Alibaba
Released
2025-04-28
Intelligence
15.3
Artificial Analysis Index
Coding
11.0
Artificial Analysis Index
In $/1M
$0.09
input tokens
Out $/1M
$0.45
output tokens
Blended $/1M
$0.18
3:1 blended
Speed
70
tokens / sec
Benchmark breakdown
Independent evaluation scores, normalized to 0–100.
GPQA Diamond
62
Humanity’s Last Exam
7
MMLU-Pro
78
SciCode
28
LiveCodeBench
51
MATH-500
96
AIME 2025
72
τ²-Bench (agentic)
26
Terminal-Bench Hard
2
IFBench
41