claude-3-7-sonnet-20250219

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: 68.6
benchmarksof 12: 8
per 1,000 documentsmean over its benchmarks: $21
seconds per inputmean response time: 14.4

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

By benchmark

Benchmark	Score	Rank	Per 1,000 docs	s / input	Best run
Bibliographic Dataolder run	92.7	–	$35$0.18 run	27.3	T0031 · 30 Sep 2025prompt.txt
Blacklist Cardsolder run	86.2	–	$8.92$0.29 run	4.8	T0320 · 24 Oct 2025prompt.txt
Book Advert XML Repair	33.5±8.4	–	$15$0.73 run	9.85	T0463 · 8 Dec 2025prompt.txt
Business Lettersolder run	45.0	–	–– run	3.47	T0017 · 11 Apr 2025prompt.txt
Company Listsolder run	72.1	–	$23$0.34 run	16.41	T0369 · 28 Oct 2025prompt.txt
Fraktur Advertsolder run	66.2	–	$36$0.18 run	39.32	T0092 · 30 Sep 2025prompt_optimized.txt
Library Cardsolder run	87.3	–	–– run	5.66	T0144 · 2 Sep 2025prompt.txt
Medieval Manuscriptsolder run	65.8	–	$11$0.13 run	8.45	T0274 · 24 Oct 2025prompt.txt

Not run on: Duty Rosters, General Meeting Minutes, Magazine Pages, Personnel Cards.

Price

From	Input, $ per million tokens	Output, $ per million tokens	Source
0 Sep 2025	3.0	15.0	archived price page

claude-3-7-sonnet-20250219

By benchmark

Price

Every run