NEWHow AI models draw the world map · Gecko Tests
Beta
Home/Models/Hermes 2 Pro - Llama-3 8B
nousresearch logo

Hermes 2 Pro - Llama-3 8B

by nousresearch · Released May 2024

Open Source
37.8
avg score
Rank #206
Compare
Better than 34% of all models
Context
8K tokens (~4 books)
Input $/1M
$0.14
Output $/1M
$0.14
Type
text
License
Open Source
Benchmarks
6 tested
Data as of
About

Hermes 2 Pro is an upgraded, retrained version of Nous Hermes 2, consisting of an updated and cleaned version of the OpenHermes 2.5 Dataset, as well as a newly introduced...

Tested on 6 benchmarks · BenchGecko score 37.8. Top scores: IFEval (53.6%), BBH (HuggingFace) (30.7%), MMLU-PRO (22.8%).

Looking for similar performance at lower cost?
GPT-5 Nano scores 37.7 (100% as good) at $0.05/1M input · 64% cheaper
Capabilities
reasoning
11.3
#170 globally
math
8.4
#257 globally
knowledge
14.3
#279 globally
language
53.6
#121 globally
general
30.7
#122 globally
Benchmark Scores
Compare All
Tested on 6 benchmarks · Ranked across 5 categories
Score Distribution (all 312 models)
0255075100
▲ You are here
reasoningCompare reasoning →
MUSR

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

11.3·
mathCompare math →
MATH Level 5

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

8.4·
knowledgeCompare knowledge →
MMLU-PRO

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

22.8·
GPQA

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

5.7·
Quick compare:Anthropic logovs Claude 2OpenAI logovs GPT-5 NanoOpenAI logovs GPT-4.1
Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Similar Models
Anthropic logo
Claude 2
Anthropic
37.7n/a
OpenAI logo
GPT-5 Nano
OpenAI
37.7$0.05/1M
OpenAI logo
GPT-4.1
OpenAI
37.9$2.00/1M
Links
Info
nousresearchPricing explorerDevelopers · API
Research
Technical Report
Documentation
API DocsPlayground
Community
@nousresearch
Source Code
GitHubHugging Face
BenchGecko API
hermes-2-pro-llama-3-8b
Specifications
  • Typetext
  • Context8K tokens (~4 books)
  • ReleasedMay 2024
  • LicenseOpen Source
  • StatusActive
  • Cost / Message~$0.000
Available On
nousresearch logonousresearch$0.14
Categories
reasoningmathknowledgelanguagegeneral
Learn More
transformeropen weightstokens
Share & Export
Tweet
Related Models
Anthropic logoClaude 2
37.7
OpenAI logoGPT-5 Nano
37.7
OpenAI logoGPT-4.1
37.9
z-ai logoGLM 4 32B
37.5
Hermes 2 Pro - Llama-3 8B is an open-source text AI model by nousresearch, released in May 2024. It has an average benchmark score of 37.8. Context window: 8K tokens.

Related Models

Claude 2 · AnthropicGPT-5 Nano · OpenAIGPT-4.1 · OpenAIGLM 4 32B · z-aiGemini 2.5 Flash · Google DeepMind

Benchmarks

IFEvalBBH (HuggingFace)MMLU-PROMUSRMATH Level 5

Related Pages

nousresearch · Providernousresearch · EconomyAll ModelsCompare ModelsPricingDevelopers · API

Key facts · as of 2026-05-03

  • Hermes 2 Pro - Llama-3 8B by nousresearch. BenchGecko score 37.8, rank 206 of 312 scored models (normalized average of public benchmark scores).
  • List price $0.14 input · $0.14 output per 1M tokens (as of 2026-05-03).

How to cite · data as of 2026-05-03

Hermes 2 Pro - Llama-3 8B · benchmarks, pricing and providers. BenchGecko, data as of 2026-05-03. https://benchgecko.ai/model/hermes-2-pro-llama-3-8b

Credit "Source: BenchGecko" with a link. Prices per provider and Gecko Tests are BenchGecko data (CC BY 4.0); benchmark scores keep their original source, listed in the JSON. JSON · llms.txt · MCP