Use Hermes Agent with OpenRouter — Multi-Model Access
Connect Hermes Agent to OpenRouter for access to 100+ models including Claude, GPT-4, and Llama.
OpenRouter gives you access to 200+ models — Claude, GPT-4, Llama, Mistral, and more — through a single API key. It's the fastest way to try different models with Hermes and add automatic fallback when your primary model is down.
Before you start:
- ☑Hermes Agent installed
- ☑An OpenRouter account at openrouter.ai (free to sign up)
- ☑An OpenRouter API key from the dashboard
- ☑Credit added to your OpenRouter account (most models cost fractions of a cent per request)
Steps
- 1
Get an OpenRouter API key
Sign up at openrouter.ai and generate an API key from the dashboard
- 2
Configure Hermes
Set model: provider: openrouter and model: apiKey: your-key in config.yaml
- 3
Choose your model
Set model: name: to any supported model like nousresearch/hermes-3-llama-3.1-405b
- 4
Set fallback models
Configure model: fallback: with backup models in case primary is unavailable
- 5
Start Hermes
hermes start — requests route through OpenRouter to your chosen model
Pro Tips
- 💡Set a spending limit in OpenRouter's dashboard to prevent unexpected bills — Hermes can be chatty with tools
- 💡Use 'nousresearch/hermes-3-llama-3.1-405b' as your primary model for best tool-use performance with Hermes Agent
- 💡Configure fallback models in config.yaml in case your primary is down or rate-limited
Troubleshooting
❌ API key works in curl but fails in Hermes
✅ Check for invisible characters or newlines in config.yaml. Use 'hermes config set model.apiKey YOUR_KEY' instead of editing the file directly.
❌ Models return 402 Payment Required
✅ Your OpenRouter balance is depleted. Add credits at openrouter.ai/credits. Some models like large Claude variants have higher per-token costs.
❌ Rate limit errors during heavy use
✅ Set 'model: rateLimit: 20' in config.yaml to limit requests per minute, or upgrade your OpenRouter tier for higher rate limits.