Skip to content
EN
English 简体中文 soon 日本語 soon

Nous Hermes

Open-weight Hermes model family

Visit official site

What Nous Hermes is

Nous Hermes is a family of open-weight models from Nous Research, built on Llama architectures and instruction-tuned for controllability, with function calling, structured outputs and long-context variants across parameter sizes.

What you can do with it

  • Self-host capable open models
  • Build agents with function calls
  • Generate structured outputs
  • Compare tuning directions

Who it is for

  • Open-source model practitioners
  • Teams needing controllable outputs
  • Researchers comparing models

What to watch out for

  • Self-hosting demands hardware and MLOps capability
  • Licence terms apply per model version
  • Benchmark claims need your own validation
  • Model lineup evolves quickly

Pros & cons

✓ What we like

  • Strong open-weight controllability
  • Function calling and structured output
  • Broad size range

! What to watch out for

  • Self-hosting effort
  • Licence review needed
  • Benchmark claims to validate

FAQ

What is Nous Hermes?

A family of open-weight Llama-based models from Nous Research with engineering-oriented features.

Does it support function calling?

Yes. Function calls and structured outputs are supported for integration.

Who suits it?

Developers and researchers needing controllable open-weight deployment.

Last reviewed: 2026-09-15

More AI chatbot tools

View all →

How we review