Creator

Thinking Machines Lab

Released

2026-07-15

Intelligence

40.7

Artificial Analysis Index

Coding

52.1

Artificial Analysis Index

In $/1M

$1.87

input tokens

Out $/1M

$4.68

output tokens

Blended $/1M

$2.57

3:1 blended

Speed

85

tokens / sec

Benchmark breakdown

Independent evaluation scores, normalized to 0–100.

GPQA Diamond
87
Humanity’s Last Exam
30
MMLU-Pro
SciCode
46
LiveCodeBench
MATH-500
AIME 2025
τ²-Bench (agentic)
Terminal-Bench Hard
IFBench
Via Artificial Analysis
© 2026 NYSGPT2525 LLC