Skip to main content
Released in February 2026 by Nous Research, Hermes Agent represents a shift toward a truly autonomous AI Agent. Unlike IDE-tethered coding copilots or simple chatbot wrappers, Hermes lives on your server, remembers what it learns, and gets more capable the longer it runs. In this guide, we’ll walk you through how to connect SiliconFlow’s APIs to Hermes Agent, so you can power your agent with 100+ SOTA AI models in just a few steps.

​
Prerequisites

​
Get Your SiliconFlow API Key

Before you begin, ensure you have a valid SiliconFlow account: You will need this key when configuring Hermes Agent.

​
Step 1: Install Hermes Agent

Hermes Agent can be installed using a single command. Run the following command in your terminal: 
curl -fsSL https://raw.githubusercontent.com/NousResearch/hermes-agent/main/scripts/install.sh | bash
Once the installation completes, the Hermes Setup Wizard will launch automatically. Press Enter to continue.
Hermes1refined

​
Step 2: Configure SiliconFlow APIs

  • Select your inference provider: Custom OpenAI-compatible endpoint (self-hosted/VLLM/ect.)
  • API base URL: http://api.siliconflow.com/v1
  • API Key: Paste Your SiliconFlow API Key
  • Model name: Enter the model you want to use. You can select from the models available in the SiliconFlow model library.
After entering the required information, press Enter to save your configuration.
Hermes2refined
Hermes3refined
Hermes Agent will continue guiding you through additional configuration steps, including:
  • TTS Provider (Text-to-Speech)
  • Terminal Backend
  • Agent Behavior Settings
These options can be configured based on your own environment and use case. If you’re unsure, you can keep the default settings and modify them later.
Hermes4refined
Hermes5refined

​
Step 3: Start Using Hermes Agent

Once the configuration is complete, start Hermes by running: hermes You’ll see the Hermes Agent interface launch in your terminal, showing available tools and skills. Type your message and press Enter to start chatting. At the bottom of the screen, you can see which model is currently active, along with token usage and response time — confirming your SiliconFlow API is connected and working.
Hermes6refined

​
Quick Reference: Useful Hermes Commands

CommandDescription
hermesStart a new chat session
hermes “your message”Send a single message
hermes -cResume your last session
hermes modelSwitch to a different model
hermes configView your current config
hermes toolsManage available tools
hermes updateUpdate to the latest version
hermes doctorDiagnose connection issues

​
Resources

​
Hermes Agent

​
SiliconFlow