Creator
Alibaba
Released
2026-04-20
Intelligence
51.8
Artificial Analysis Index
Coding
44.9
Artificial Analysis Index
In $/1M
$1.30
input tokens
Out $/1M
$7.80
output tokens
Blended $/1M
$2.92
3:1 blended
Speed
41
tokens / sec
Benchmark breakdown
Independent evaluation scores, normalized to 0–100.
GPQA Diamond
89
Humanity’s Last Exam
29
MMLU-Pro—
SciCode
47
LiveCodeBench—
MATH-500—
AIME 2025—
τ²-Bench (agentic)
96
Terminal-Bench Hard
44
IFBench
77