← All benchmarks

Claude Opus 5 (Adaptive Reasoning, Low Effort)

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

Intelligence Index
50.6
Output speed
59.7
tokens/sec
Blended price
$10.00
per 1M tokens
Time to first token
3.01s

Model metadata

Model ID20928ba9-3a3f-415f-9519-b84ff64ecf34
Slugclaude-opus-5-low
Release date2026-07-24
ProviderAnthropic
Provider sluganthropic
Output speed59.7 tok/s
Time to first token3.01s
Time to first answer3.01s
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 index66.90
artificial analysis intelligence index50.60
artificial analysis math index
gpqa88.9%
hle41.3%
ifbench
lcr69.3%
livecodebench
math 500
mmlu pro
scicode48.0%
tau banking23.3%
tau2
terminalbench hard
terminalbench v2 176.4%

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

Benchmark data from Artificial Analysis.