Hermes 4 405B

Hermes 4 405B places in the top three on 2 benchmarks.

Nous Research Released 26 Aug 2025 Family Hermes 4 The lab's model card

Benchmarks with a score

Ranked first, out of five or more models

In the top three

Nous Research Aug 2025

Open weights Flagship

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
Arena-Hard-Auto	94.4%	#3 of 39	Model card
OpenBookQA	94.2%	#2 of 20	Model card
BIG-Bench Hard	86.3%	#20 of 71	Model card
MMLU	87.2%	#38 of 127	Model card
EQ-Bench 3	1,405	#33 of 78	Official leaderboard
DROP	83.5%	#17 of 40	Model card
MMLU-Pro	80.6%	#71 of 151	Model card
SimpleQA	25.8%	#28 of 59	Model card
Creative Writing v3	66.7	#63 of 120	Official leaderboard
MuSR	66.1%	#4 of 6	Model card
Longform Creative Writing (EQ-Bench)	51.2	#81 of 119	Official leaderboard
IFEval	81.5%	#94 of 126	Model card
RewardBench	73%	#2 of 2	Model card

Benchmark	Score	Rank	Source
AIME 2024	81.9%	#35 of 87	Model card
MATH-500	96.3%	#30 of 67	Model card
GPQA Diamond	70.6%	#170 of 284	Model card
AIME 2025	78.1%	#91 of 143	Model card

Benchmark	Score		Rank	Source
LiveCodeBench v6	61.4%		#64 of 76	Model card

Benchmark	Score		Rank	Source
SnakeBench	21		#123 of 183	Official leaderboard

Benchmark	Score		Rank	Source
RefusalBench (Nous Research)	57.1%		#2 of 2	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.