Creator
Alibaba
Released
2025-04-28
Intelligence
12.5
Artificial Analysis Index
Coding
13.3
Artificial Analysis Index
In $/1M
$0.08
input tokens
Out $/1M
$0.29
output tokens
Blended $/1M
$0.13
3:1 blended
Speed
72
tokens / sec
Benchmark breakdown
Independent evaluation scores, normalized to 0–100.
GPQA Diamond
52
Humanity’s Last Exam
5
MMLU-Pro
71
SciCode
26
LiveCodeBench
32
MATH-500
86
AIME 2025
22
τ²-Bench (agentic)
22
Terminal-Bench Hard
7
IFBench
32