Claude Opus 4.8

Adaptive Reasoning, Max Effort

Creator

Anthropic

Released

2026-05-28

Intelligence

61.4

Artificial Analysis Index

Coding

56.7

Artificial Analysis Index

In $/1M

$6.25

input tokens

Out $/1M

$25.00

output tokens

Blended $/1M

$10.94

3:1 blended

Speed

69

tokens / sec

Benchmark breakdown

Independent evaluation scores, normalized to 0–100.

GPQA Diamond
92
Humanity’s Last Exam
46
MMLU-Pro
SciCode
54
LiveCodeBench
MATH-500
AIME 2025
τ²-Bench (agentic)
94
Terminal-Bench Hard
58
IFBench
62
Via Artificial Analysis