modelgrep
N

Nous: Hermes 4 70B

nousresearch/hermes-4-70b

138th smartest of 179Cheaper than 76% of paidReasoningJSON
Use via OpenRouter ↗
Intelligence
16.0
138th of 179
Design Elo
Speed
67
113th fastest
Latency
252ms
first token
Input price
$0.130
71st cheapest
Context
131K

How it compares

Smarter than23%
of all ranked models
Faster than63%
of all ranked models
Cheaper than76%
of all ranked models

Overview

Hermes 4 70B is a hybrid reasoning model from Nous Research, built on Meta-Llama-3.1-70B. It introduces the same hybrid mode as the larger 405B release, allowing the model to either...

Benchmarks

independent · via OpenRouter
Artificial Analysis22th percentile
Intelligence Index
16.0
Coding Index
14.4
Agentic Index
11.7
GPQA Diamond
70%
Humanity's Last Exam
8%
SciCode
34%
Tau²-Bench (agentic)
23%

Providers & pricing (1)

ProviderIn $/MOut $/MUptime
Nebiusfp8$0.130$0.400100%

Specifications

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

Hermes 4 70B FAQ

How much does Hermes 4 70B cost?

Hermes 4 70B costs $0.130 per million input tokens and $0.400 per million output tokens via OpenRouter, making it 71st cheapest of 298 paid models.

How smart is Hermes 4 70B?

Hermes 4 70B scores 16.0 on the Artificial Analysis Intelligence Index, ranking 138th of 179 benchmarked models, with a GPQA Diamond score of 70%.

How fast is Hermes 4 70B?

Hermes 4 70B generates around 67 tokens per second with 252ms time-to-first-token (p50), the 113th fastest tracked model.

What is Hermes 4 70B's context window?

Hermes 4 70B supports a 131K-token context window. It accepts text input.

Compare head-to-head