How much does Hermes 3 70B Instruct cost?

Hermes 3 70B Instruct costs $0.30 per million input tokens and $0.30 per million output tokens. For a typical conversation (~2,000 tokens), that's approximately $0.001 per message.

What benchmarks has Hermes 3 70B Instruct been tested on?

Hermes 3 70B Instruct has been evaluated on 6 benchmarks. Top scores: IFEval: 76.6, BBH (HuggingFace): 53.8, MMLU-PRO: 41.4.

Is Hermes 3 70B Instruct open source?

Yes, Hermes 3 70B Instruct is open source.

How does Hermes 3 70B Instruct compare to Gemini 2.5 Pro Preview 06-05?

Hermes 3 70B Instruct has an average score of 73.3 while Gemini 2.5 Pro Preview 06-05 scores 73.5. Gemini 2.5 Pro Preview 06-05 slightly outperforms Hermes 3 70B Instruct overall. Hermes 3 70B Instruct costs $0.30/1M input vs Gemini 2.5 Pro Preview 06-05 at $1.25/1M input. See full comparison →

Home/Models/Hermes 3 70B Instruct

Hermes 3 70B Instruct

Name: Hermes 3 70B Instruct
Price: 0.3 USD
Author: nousresearch

by nousresearch · Released Aug 2024

Open Source

73.3

avg score

Rank #33

Compare

Better than 86% of all models

Context

131K tokens (~66 books)

Input $/1M

$0.30

Output $/1M

$0.30

Type

text

License

Open Source

Benchmarks

6 tested

Data updated today

About

Hermes 3 is a generalist language model with many improvements over [Hermes 2](/models/nousresearch/nous-hermes-2-mistral-7b-dpo), including advanced agentic capabilities, much better roleplaying, reasoning, multi-turn conversation, long context coherence, and improvements across the...

Tested on 6 benchmarks with 38.5% average. Top scores: IFEval (76.6%), BBH (HuggingFace) (53.8%), MMLU-PRO (41.4%).

Looking for similar performance at lower cost?
gpt-oss-120b (free) scores 74.2 (101% as good) at $0.00/1M input · 100% cheaper

Capabilities

reasoning

23.4

#98 globally

math

21.0

#157 globally

knowledge

28.1

#180 globally

language

76.6

#55 globally

general

53.8

#7 globally

Benchmark Scores

Compare All

Tested on 6 benchmarks · Ranked across 5 categories

Score Distribution (all 233 models)

0255075100

▲ You are here

reasoningCompare reasoning →

MUSR

HuggingFace MuSR (Multi-Step Reasoning). Tests multi-hop reasoning requiring chaining multiple facts together.

23.4—

mathCompare math →

MATH Level 5

HuggingFace evaluation of MATH Level 5 problems. Competition math requiring advanced reasoning and proof construction.

21.0—

knowledgeCompare knowledge →

MMLU-PRO

HuggingFace MMLU-Pro. Harder version of MMLU with 10 answer choices instead of 4 and more challenging questions.

41.4—

GPQA

HuggingFace evaluation of GPQA (Graduate-Level Google-Proof Q&A). PhD-level science questions that cannot be easily searched.

14.9—

Quick compare:

vs Gemini 2.5 Pro Preview 06-05

vs MiniMax M2

vs gpt-oss-120b (free)

Excellent (85+) Good (70-85) Average (50-70) Below (<50)

Similar Models

Gemini 2.5 Pro Preview 06-05

Links

Info

nousresearch Pricing explorer Developers · API

Research

Documentation

Community

Source Code

BenchGecko API

hermes-3-llama-3-1-70b

Specifications

Typetext
Context131K tokens (~66 books)
ReleasedAug 2024
LicenseOpen Source
StatusActive
Cost / Message~$0.001

Available On

nousresearch$0.30

Frequently Asked Questions

Hermes 3 70B Instruct is an open-source text AI model by nousresearch, released in August 2024. It has an average benchmark score of 73.3. Context window: 131K tokens.

Benchmarks

IFEval BBH (HuggingFace)MMLU-PRO MUSR MATH Level 5

nousresearch · Provider nousresearch · Economy All Models Compare Models Pricing Developers · API Context Window · Glossary

Hermes 3 70B Instruct

Frequently Asked Questions

Related Models

Benchmarks

Related Pages