Skip to content

claude-opus-4-20250514

Anthropic

legacy model Its test configurations are retired, so it is no longer run. The figures are from its best legacy runs, scored as they ran at the time; it is not ranked.

average: 62.8
benchmarksof 12: 10
per 1,000 documentsmean over its benchmarks: $105
seconds per inputmean response time: 21.8

Compare with other models Its row in the Matrix All its test runs

By task

Data correction

50.5

Best of 114 models with all its benchmarks: claude-opus-5-5, 82.7.
Transcription

69.4

Best of 114 models with all its benchmarks: gemini-4-argon, 96.2.

By benchmark

Benchmark	Score	Rank	Per 1,000 docs	s / input	Best run
Bibliographic Data	94.2±2.4	–	$179$0.90 run	35.7	T0106 · 24 Jan 2026prompt.txt
Blacklist Cards	90.7±3.1	–	$52$1.73 run	6.91	T0322 · 24 Jan 2026prompt.txt
Book Advert XML Repair	25.0±7.5	–	$71$3.56 run	13.72	T0465 · 24 Jan 2026prompt.txt
Business Letters	46.0±6.7	–	$63$3.58 run	4.87	T0100 · 24 Jan 2026prompt.txt
Company Listsolder run	78.7	–	$113$1.70 run	24.45	T0373 · 28 Oct 2025prompt.txt
Fraktur Adverts	58.1±46.5	–	$234$1.17 run	87.72	T0098 · 25 Jan 2026prompt_optimized.txt
Library Cardsolder run	82.8	–	–– run	6.12	T0147 · 2 Sep 2025prompt.txt
Magazine Pages	2.2±4.3	–	$39$1.82 run	4.34	T0775 · 23 Mar 2026prompt.txt
Medieval Manuscripts	74.1±13.5	–	$62$0.75 run	15.81	T0290 · 24 Jan 2026prompt.txt
Personnel Cards	76.0±11.1	–	$134$8.19 run	18.43	T0578 · 9 Feb 2026prompt.txt

Not run on: Duty Rosters, General Meeting Minutes.

Price

From	Input, $ per million tokens	Output, $ per million tokens	Source
0 Sep 2025	15.0	75.0	archived price page

Every run