Hermes Agent

AI agent with learning capabilities, able to accumulate skills from execution experience and autonomously optimize its abilities

⚡ Nous Research

📖 Introduction

Hermes Agent is an open-source AI agent developed by Nous Research. It's a rare AI agent with a natively built-in learning loop, capable of accumulating skills from execution experience, autonomously optimizing capabilities, persisting knowledge, retrieving historical conversations, and continuously refining user cognitive models across sessions.

Hermes supports multiple LLM providers and provides rich tool integration capabilities.

⚙️ System Requirements

Component Requirement Notes
Python Python 3.10+ Conda or venv recommended
API Key Supports multiple LLM providers OpenAI / Anthropic / Google, etc.
Memory 8GB+ recommended Depends on usage scenario

🖥️ Choose Your Platform

Install via pip

Terminal
# Create virtual environment (recommended) python -m venv hermes-env source hermes-env/bin/activate # Install Hermes Agent pip install hermes-agent # Or install latest version from GitHub pip install git+https://github.com/NousResearch/hermes-agent.git

Run Hermes

Terminal
# Start interactive session hermes model # Configure tools hermes tools # View help hermes --help

Configure LLM Provider

Terminal
# Configure via environment variables export OPENAI_API_KEY="your-api-key" export OPENAI_BASE_URL="https://api.beehears.com/v1" # Or configure Anthropic export ANTHROPIC_API_KEY="your-api-key" export ANTHROPIC_BASE_URL="https://api.beehears.com" # Run hermes model

Install via pip

PowerShell or CMD
# Create virtual environment python -m venv hermes-env .\hermes-env\Scripts\activate # Install Hermes Agent pip install hermes-agent

Install via WSL

WSL Terminal
# Create virtual environment python3 -m venv hermes-env source hermes-env/bin/activate # Install pip install hermes-agent # Run hermes model

Configure Environment Variables (PowerShell)

PowerShell
$env:OPENAI_API_KEY="your-api-key" $env:OPENAI_BASE_URL="https://api.beehears.com/v1" hermes model

Configure Environment Variables (CMD)

CMD
set OPENAI_API_KEY=your-api-key set OPENAI_BASE_URL=https://api.beehears.com/v1 hermes model

Install via pip

Terminal
# Install Python and pip (if not installed) sudo apt update sudo apt install python3 python3-pip python3-venv # Create virtual environment python3 -m venv hermes-env source hermes-env/bin/activate # Install Hermes Agent pip install hermes-agent

Install from Source

Terminal
# Clone repository git clone https://github.com/NousResearch/hermes-agent.git cd hermes-agent # Create virtual environment and install python3 -m venv venv source venv/bin/activate pip install -e .

Configure and Run

Terminal
# Configure environment variables export OPENAI_API_KEY="your-api-key" export OPENAI_BASE_URL="https://api.beehears.com/v1" # Run interactive CLI hermes model # Configure tools hermes tools
💡 Using BeeHears

Through BeeHears API, you can access multiple models including GPT, Claude, and Gemini series.

🚀 Basic Usage

Main Commands

  • hermes model - Start interactive CLI, select a model to begin conversation
  • hermes tools - Configure available tools
  • hermes --help - View help information

Core Features

  • Learning Loop - Accumulate skills and knowledge from execution experience
  • Multi-provider Support - Supports multiple LLM providers
  • Tool Integration - Rich tool extension capabilities
  • History Retrieval - Retrieve historical conversations, continuously optimize across sessions

Using with OpenClaw

Hermes Agent can be used alongside OpenClaw to build more powerful AI automation solutions. OpenClaw provides Gateway and multi-channel integration capabilities, while Hermes provides intelligent agent and learning capabilities.

🔗 Related Links