Nova 2 Omni Preview

Nova 2 Omni Preview ranks first on 2 of the 16 benchmarks it has scores on.

Amazon Released 2 Dec 2025 Family Amazon Nova 2 The lab's model card

Benchmarks with a score

Ranked first, out of five or more models

In the top three

Amazon Dec 2025

Closed weights Mid-size

No Epoch Capabilities Index score yet

Where it ranks

Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.

Benchmark	Score	Rank	Source
MultiChallenge	75.5%	#3 of 36	Model card
MMLU-Pro	80.7%	#69 of 151	Model card
IFBench	68.7%	#46 of 70	Model card

Benchmark	Score		Rank	Source
AIME 2025	92.1%		#41 of 143	Model card

Benchmark	Score	Rank	Source
tau2-bench Verified	73.5%	#1 of 3	Model card
tau2-bench	80%	#35 of 69	Model card
BFCL V4	58.3%	#18 of 24	Model card
ScreenSpot	85.4%	#9 of 11	Model card

Benchmark	Score	Rank	Source
CoVoST2	40.7	#1 of 11	Model card
MMAU v05.15.25	77.8%	#1 of 9	Official leaderboard
MMAU	75.3%	#8 of 14	Model card
QVHighlights	76.7%	#3 of 5	Model card
Common Voice	5.70%	#2 of 3	Model card
Video-MME	77.9%	#30 of 42	Model card
OCRBench v2	58.2%	#15 of 20	Model card
MMMU-Pro	61.4%	#55 of 72	Model card
RefCOCO / RefCOCO+ / RefCOCOg	86.3%	#18 of 19	Model card
RealKIE-FCC Verified	59.8%	#3 of 3	Model card
ImgEdit-Bench	4.12	#1 of 1	Model card
MAVERIX	66.6%	#1 of 1	Model card
Multilingual LibriSpeech	4.70	#1 of 1	Model card

Benchmark	Score	Rank	Source
Jigsaw Toxic Comment Classification	57.8%	#1 of 3	Model card
WildGuardTest	84.1%	#5 of 5	Model card
Aegis 2.0	80.7%	#3 of 3	Model card

The benchmarks the lab chose to headline, with the scores it gave.

Every result in the longer documents the lab published with it. 4 of them aren't in the launch post.