Skip to content

command-a-vision-07-2025

Cohere

average±4.1 (95%): 40.7
leaderboard rankof 114: ≈110
benchmarksof 12: 12
per 1,000 documentsmean over its benchmarks: $15
seconds per inputmean response time: 100.6

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

By task

Data correction

39.6

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

42.7

Best of 107 models with all its benchmarks: gpt-6-astra, 97.0.
Information extraction

52.5

Best of 107 models with all its benchmarks: grok-4.6, 90.8.
Transcription

39.6

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	39.3±26.5	≈110of 115	–$0 run	401.76	T1141 · 17 Sep 2026prompt.txt
Blacklist Cards	81.4±8.8	≈99of 114	–$0 run	4.2	T1148 · 18 Sep 2026prompt.txt
Book Advert XML Repair	0.7±0.7	≈114of 119	–$0 run	1.81	T0487 · 17 Sep 2026prompt.txt
Business Letters	35.0±5.8	112of 114	–$0 run	5.17	T1142 · 8 Sep 2026prompt.txt
Company Lists	41.9±19.9	≈105of 114	–$0 run	24.39	T1149 · 18 Sep 2026prompt.txt
Duty Rosters	48.4±7.2	≈96of 107	–$0 run	57.84	T1280 · 18 Sep 2026prompt.txt
Fraktur Adverts	6.4±12.6	≈104of 114	–$0 run	365.1	T1145 · 17 Sep 2026prompt_optimized.txt
General Meeting Minutes	59.6±23.7	≈85of 108	–$0 run	17.78	T1220 · 18 Sep 2026prompt.txt
Library Cards	61.9±2.3	108of 114	–$0 run	62.3	T1146 · 17 Sep 2026prompt.txt
Magazine Pages	1.1±2.1	≈79of 115	–$0 run	211.65	T0802 · 17 Sep 2026prompt.txt
Medieval Manuscripts	34.0±18.0	≈95of 114	–$0 run	33.46	T1147 · 18 Sep 2026prompt.txt
Personnel Cards	78.5±9.9	≈94of 114	$15$0.91 run	21.57	T0582 · 10 Feb 2026prompt.txt

Price

From	Input, $ per million tokens	Output, $ per million tokens	Source
2 Mar 2026	0.0	0.0	archived price page
9 Dec 2025	2.5	10.0	archived price page

Every run