Skip to content

gpt-5.2-2025-12-11

OpenAI

average±2.2 (95%): 72.7
leaderboard rankof 114: ≈54
benchmarksof 12: 12
per 1,000 documentsmean over its benchmarks: $15
seconds per inputmean response time: 34.9

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

By task

Data correction

62.1

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

89.8

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

71.1

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

85.9

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±5.0	≈57of 115	$33$0.17 run	27.74	T0488 · 23 Mar 2026prompt.txt
Blacklist Cards	79.9±5.4	≈99of 114	$5.55$0.18 run	3.85	T0495 · 10 Feb 2026prompt.txt
Book Advert XML Repair	28.0±7.6	≈41of 119	$9.12$0.46 run	8.89	T0498 · 10 Feb 2026prompt.txt
Business Letters	58.0±7.0	≈24of 114	$12$0.70 run	2.92	T0489 · 16 Sep 2026prompt.txt
Company Lists	67.7±13.8	≈84of 114	$16$0.24 run	7.73	T0496 · 16 Sep 2026prompt.txt
Duty Rosters	84.6±3.7	≈18of 107	$21$0.44 run	11.99	T1302 · 16 Sep 2026prompt.txt
Fraktur Adverts	90.6±3.9	≈32of 114	$28$0.14 run	26.86	T0492 · 16 Sep 2026prompt_optimized.txt
General Meeting Minutes	80.8±11.7	≈25of 108	$16$0.15 run	8.56	T0623 · 12 Mar 2026prompt.txt
Library Cards	32.6±4.9	112of 114	$4.21$1.11 run	300.59	T0493 · 26 Jan 2026prompt.txt
Magazine Pages	88.5±7.1	≈18of 115	$4.13$0.19 run	3.48	T0803 · 23 Mar 2026prompt.txt
Medieval Manuscripts	70.9±11.3	≈27of 114	$10$0.12 run	7.0	T0494 · 16 Sep 2026prompt.txt
Personnel Cards	96.2±1.5	≈30of 114	$16$0.99 run	9.38	T0614 · 9 Feb 2026prompt.txt

Price

From	Input, $ per million tokens	Output, $ per million tokens	Source
16 Mar 2026	1.75	14.0	archived price page

Every run