← All benchmarks

Claude Opus 5 (Adaptive Reasoning, Xhigh Effort)

Anthropic · Released 2026-07-24 · Independent benchmark scores from Artificial Analysis

Intelligence Index
60.1
Output speed
56.9
tokens/sec
Blended price
$10.00
per 1M tokens
Time to first token
22.36s

Model metadata

Model ID1305c921-7aaa-4d6d-99b5-99b3acf15e19
Slugclaude-opus-5-xhigh
Release date2026-07-24
ProviderAnthropic
Provider sluganthropic
Output speed56.9 tok/s
Time to first token22.36s
Time to first answer22.36s
Blended price ($/1M)$10.00
Input price ($/1M)$5.00
Output price ($/1M)$25.00

All evaluation scores (17)

aime
aime 25
artificial analysis coding index77.00
artificial analysis intelligence index60.10
artificial analysis math index
gpqa93.7%
hle52.5%
ifbench
lcr68.7%
livecodebench
math 500
mmlu pro
scicode55.0%
tau banking31.5%
tau2
terminalbench hard
terminalbench v2 188.0%

Source: artificialanalysis.ai/models/claude-opus-5-xhigh

Benchmark data from Artificial Analysis.