Creator
Alibaba
Released
2026-04-16
Intelligence
24.2
Artificial Analysis Index
Coding
28.1
Artificial Analysis Index
In $/1M
$0.38
input tokens
Out $/1M
$2.25
output tokens
Blended $/1M
$0.84
3:1 blended
Speed
101
tokens / sec
Benchmark breakdown
Independent evaluation scores, normalized to 0–100.
GPQA Diamond
82
Humanity’s Last Exam
13
MMLU-Pro—
SciCode
1
LiveCodeBench—
MATH-500—
AIME 2025—
τ²-Bench (agentic)
85
Terminal-Bench Hard
26
IFBench
36