Creator
Alibaba
Released
2026-03-30
Intelligence
25.9
Artificial Analysis Index
Coding
14.0
Artificial Analysis Index
In $/1M
$0.10
input tokens
Out $/1M
$0.80
output tokens
Blended $/1M
$0.28
3:1 blended
Speed
286
tokens / sec
Benchmark breakdown
Independent evaluation scores, normalized to 0–100.
GPQA Diamond
74
Humanity’s Last Exam
7
MMLU-Pro—
SciCode
26
LiveCodeBench—
MATH-500—
AIME 2025—
τ²-Bench (agentic)
85
Terminal-Bench Hard
8
IFBench
38