modelgrep
N

Nous: Hermes 4 405B

nousresearch/hermes-4-405b

128th smartest of 178ReasoningJSON
Use via OpenRouter ↗
Intelligence
18.6
128th of 178
Design Elo
Speed
32
230th fastest
Latency
465ms
first token
Input price
$1.00
207th cheapest
Context
131K

How it compares

Smarter than28%
of all ranked models
Faster than22%
of all ranked models
Cheaper than31%
of all ranked models

Overview

Hermes 4 is a large-scale reasoning model built on Meta-Llama-3.1-405B and released by Nous Research. It introduces a hybrid reasoning mode, where the model can choose to deliberate internally with...

Benchmarks

independent · via OpenRouter
Artificial Analysis28th percentile
Intelligence Index
18.6
Coding Index
16.0
Agentic Index
12.6
GPQA Diamond
73%
Humanity's Last Exam
10%
SciCode
25%
Tau²-Bench (agentic)
22%

Providers & pricing (1)

ProviderIn $/MOut $/MUptime
Nebiusfp8$1.00$3.00100%

Specifications

Context window131K
Max output
Knowledge cutoffAug 2024
Input modalitiestext
Output modalitiestext
Prompt caching
Cache read price
ModeratedNo

Hermes 4 405B FAQ

How much does Hermes 4 405B cost?

Hermes 4 405B costs $1.00 per million input tokens and $3.00 per million output tokens via OpenRouter, making it 207th cheapest of 298 paid models.

How smart is Hermes 4 405B?

Hermes 4 405B scores 18.6 on the Artificial Analysis Intelligence Index, ranking 128th of 178 benchmarked models, with a GPQA Diamond score of 73%.

How fast is Hermes 4 405B?

Hermes 4 405B generates around 32 tokens per second with 465ms time-to-first-token (p50), the 230th fastest tracked model.

What is Hermes 4 405B's context window?

Hermes 4 405B supports a 131K-token context window. It accepts text input.

Compare head-to-head