Hermes 4 - Llama-3.1 70B (Non-reasoning)

Last updated: Recently

Model Type

Text

Max Context

Max Output

Parameter Scale

Hermes 4 Llama 3.1 70B Non reasoning by Nous Research: Text model; TTFT 0.613s, 76.7 tok/s.

Input Modality

Output Modality

Inference Speed

76.742 tokens/s

Latest Release Date

8/27/2025

SDK Ecosystem

artificial-analysismanufactureraa-bootstrap

Model Overview Fields

name

Hermes 4 - Llama-3.1 70B (Non-reasoning)

Release date

8/27/2025

Performance

76.742 tokens/s

model_creator.name

Nous Research

AA Evaluation Scores

AA Intelligence Index

12.6

AA Coding Index

9.2

AA Math Index

11.3

MMLU Pro

0.664

GPQA

0.491

HLE

0.036

LiveCodeBench

0.269

SciCode

0.277

Math-500

AIME

AIME 2025

0.113

IFBench

0.29

LCR

0.02

TerminalBench Hard

TAU2

0.216

Output Speed (tokens/s)

76.742 tok/s

TTFT (s)

0.613s

First Answer Token (s)

0.613s

Nous Research

◎

Hermes 4 - Llama-3.1 70B (Non-reasoning)

$0.13/M$0.40/M

Hermes 4 Llama 3.1 70B Non reasoning by Nous Research: Text model; TTFT 0.613s, 76.7 tok/s.

There are no reviews for this model yet.

artificial-analysismanufactureraa-bootstrap

0.0