Nova 2.0 Pro Preview

Nova 2.0 Pro Preview ranks first on MultiChallenge.

Amazon Released 2 Dec 2025

Benchmarks with a score

Ranked first, out of five or more models

In the top three

Amazon Dec 2025

Closed weights

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	77.7%	#1 of 36	Model card
IFBench	80.2%	#7 of 70	Model card
MMLU-Pro	81.6%	#61 of 151	Model card
LongCodeBench	84%	#1 of 2	Model card

Benchmark	Score	Rank	Source
AIME 2025	92.3%	#40 of 143	Model card
GPQA Diamond	81.4%	#102 of 284	Model card
CritPt	0%	#147 of 160	Epoch AI

Benchmark	Score		Rank	Source
LiveCodeBench v5	74.6%		#3 of 44	Model card
SWE-bench Verified	70%		#76 of 150	Model card

Benchmark	Score	Rank	Source
tau2-bench	92.7%	#9 of 69	Model card
Terminal-Bench	41.3%	#9 of 22	Model card
ScreenSpot	88.1%	#5 of 11	Model card
BFCL V4	61.6%	#15 of 24	Model card
tau2-bench Verified	71.5%	#2 of 3	Model card

Benchmark	Score	Rank	Source
RealKIE-FCC Verified	67%	#1 of 3	Model card
OCRBench v2	64.5%	#7 of 20	Model card
QVHighlights	76.7%	#2 of 5	Model card
MMMU-Pro	63.5%	#51 of 72	Model card
GDP.pdf	2.00%	#39 of 39	Epoch AI
Common Voice	5.50%	#3 of 3	Model card

Benchmark	Score	Rank	Source
WildGuardTest	84.9%	#3 of 5	Model card
Aegis 2.0	81.4%	#2 of 3	Model card
Jigsaw Toxic Comment Classification	54.7%	#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.