hermes

hermes is an agent for coding agents from zorost/AI-Engineering-Lab. It costs 0 tokens per session (3,192 once invoked), scanned A, original, MIT.

A guide explaining that Hermes can mean either Nous Research's downloadable language-model family or its agent framework for running models.

In plain words
What is it for?
Use it to understand Hermes models and Hermes Agent, including local model use and agent setup; fast-changing details should be checked against current documentation.
Why use it?
It prevents confusion between the model, which generates responses, and the framework, which runs an agent that uses a model.

Agent

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/zorost/ai-engineering-lab/hermes
Clone the repo
git clone --depth 1 https://github.com/zorost/AI-Engineering-Lab

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for hermes

README.md
[![agentmods](https://agentmods.dev/badge/agents/zorost/ai-engineering-lab/hermes.svg)](https://agentmods.dev/agents/zorost/ai-engineering-lab/hermes)
Your own site
<a href="https://agentmods.dev/agents/zorost/ai-engineering-lab/hermes"><img src="https://agentmods.dev/badge/agents/zorost/ai-engineering-lab/hermes.svg" alt="Measured on agentmods" height="20"></a>
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 3,192 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.03192
Opus 5 $0.00000 $0.01596
Sonnet 5 $0.00000 $0.00638
Haiku 4.5 $0.00000 $0.00319

Measured 4d ago against content hash f5cfdece6e49, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

hermes scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

reference/agents/hermes.md · 255 lines

How it starts

The opening of the file, as written. The whole thing — 255 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Hermes: Model Family vs. Agent Framework

Part of AI Engineering Lab · Developed by Zorost Intelligence AI Lab · zorost.com

"Hermes" is overloaded, and getting it wrong causes real confusion. In an agents context it refers to two related-but-distinct things from Nous Research:

  1. Hermes: the open-weight model family (the "brain" you download and run).
  2. Hermes Agent: the agent framework (the runtime that uses a model).

This guide covers both, clearly labeled, then shows how to run Hermes models locally and use one as your ZoroLogistics triage-agent brain.

Note on fast-moving facts. Model sizes, base architectures, and licenses change as new generations ship. Everything marked "verify against live docs" should be checked against the model cards and repos before citing.


Part A: Hermes, the open-weight model family

Nous Research's Hermes models are instruction-tuned, open-weight LLMs strongly oriented toward instruction-following, function/tool calling, and agentic use, plus a reputation for strong roleplay/character and long-form generation.

A.1 The generations

  • Hermes 2 (early 2024), a broad family across several bases (Llama 3, Mistral, Mixtral, Yi). The "Hermes 2 Pro" line added reliable function calling and JSON/structured-output modes.
  • Hermes 3 (Aug 2024), instruction-tuned on Llama 3.1 (8B, 70B, 405B); marketed for function calling, structured outputs, roleplaying, and long context, using the ChatML format (Hermes-3-Llama-3.1-405B).
  • Hermes 4 (late 2025), the current generation: 14B and 70B, built on Qwen3 bases, licensed Apache 2.0, with a "hybrid mode" reasoning toggle and strong agentic/function-calling orientation (Hermes-4-14B, Hermes-4-70B).

Verify the current Hermes 4 sizes, base model, and license against the live model cards, these are exactly the fast-moving facts that drift between generations.

Read the full file on GitHub · 255 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 4d ago First seen · 255 lines · 0 tokens per session scan A f5cfdece6e49

Subscribe to this mod's changes

hermes is an agent published in the GitHub repository zorost/AI-Engineering-Lab (304 stars, last pushed 16d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 3,192 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.