Claude Sonnet 4.6

Adaptive Reasoning, Max Effort

Creator

Anthropic

Released

2026-02-17

Intelligence

51.7

Artificial Analysis Index

Coding

50.9

Artificial Analysis Index

In $/1M

$3.75

input tokens

Out $/1M

$15.00

output tokens

Blended $/1M

$6.56

3:1 blended

Speed

67

tokens / sec

Benchmark breakdown

Independent evaluation scores, normalized to 0–100.

GPQA Diamond
88
Humanity’s Last Exam
30
MMLU-Pro
SciCode
47
LiveCodeBench
MATH-500
AIME 2025
τ²-Bench (agentic)
76
Terminal-Bench Hard
53
IFBench
57
Via Artificial Analysis