LLM Reference

Hermes 4 Models by Nous Research

Nous ResearchLlama 3 CommunityOpen weightsOpen Source
2 models2025Up to 128k ctxFrom $0.05/1M input

Details

ResearcherNous Research
Commercial useCommercial use with conditions
Models2
Released2025
Max context128k

Capabilities

ReasoningAll models

Links

Website

About

The Hermes 4 family is Nous Research's open-source instruction-tuned series built on Llama 3.1 foundations, spanning 70B and 405B parameter variants with hybrid reasoning behavior and hosted availability through Nous Portal.

Current Variants

Use-when guidance is derived from seed capabilities, context, release, and replacement fields.

2 in view

Use when the workload needs 128k context, 70B parameters, and reasoning.

2025-09128k context70B parametersreasoning

Use when the workload needs 128k context, 405B parameters, and reasoning.

2025-09128k context405B parametersreasoning

Release Timeline

1 release group
2025-09
2 current
Hermes-4-405B
128k context405B parametersreasoning
Current
Hermes-4-70B
128k context70B parametersreasoning
Current

Specifications(2 models)

Hermes 4 model specifications comparison
ModelReleasedContextParametersReasoning
Hermes-4-70B2025-09128k70BYes
Hermes-4-405B2025-09128k405BYes

Available From(1 provider)

Pricing

Hermes 4 model pricing by provider
ModelProviderInput / 1MOutput / 1MType
Hermes-4-70BNous Portal$0.05$0.2Serverless
Hermes-4-405BNous Portal$0.09$0.37Serverless

Frequently Asked Questions

What is Hermes 4 used for?
Hermes 4 is used for reasoning. The family description and listed model capabilities point to those workloads as the best fit.
How does Hermes 4 compare to MOSS-Audio?
Hermes 4 by Nous Research is strongest where you need reasoning, while MOSS-Audio by MOSI Intelligence is the closest related family to check for multimodal. Hermes 4 has 2 listed variants and reaches up to 128k context, so compare the specs and pricing tables before choosing a production model.
Which Hermes 4 model should I use?
For the lowest listed input price, start with Hermes-4-70B through Nous Portal at $0.05/1M input tokens. For the most capable/latest local choice, evaluate Hermes-4-70B with 128k context and reasoning.