Creator
Alibaba
Released
2026-03-02
Intelligence
9.9
Artificial Analysis Index
Coding
1.0
Artificial Analysis Index
In $/1M
$0.01
input tokens
Out $/1M
$0.05
output tokens
Blended $/1M
$0.02
3:1 blended
Speed
90
tokens / sec
Benchmark breakdown
Independent evaluation scores, normalized to 0–100.
GPQA Diamond
24
Humanity’s Last Exam
5
MMLU-Pro—
SciCode
3
LiveCodeBench—
MATH-500—
AIME 2025—
τ²-Bench (agentic)
65
Terminal-Bench Hard
0
IFBench
22