Skip to main content
Last updated: 2026-06-06
Use crazyrouter-hermes to point Hermes Agent at Crazyrouter. After setup, Hermes uses https://api.crazyrouter.com/v1 as its OpenAI-compatible endpoint and calls Claude, GPT, DeepSeek, Gemini, Qwen, and 627+ other models through a single Crazyrouter API key.

​
Best For

  • Hermes Agent users who want to use Crazyrouter as the model provider
  • local agent workflows that should share Crazyrouter models and quota
  • users who want a one-command setup for OPENAI_API_KEY, OPENAI_BASE_URL, and the default model

​
Prerequisites

ItemNotes
Hermes AgentInstall Hermes locally and confirm it can start
Crazyrouter accountCreate one at crazyrouter.com
Crazyrouter API keyCreate a dedicated sk-... token for Hermes
Model accessAllow the models you plan to use in that token
Hermes connects to Crazyrouter through OpenAI-compatible settings. Use https://api.crazyrouter.com/v1 as the default access address. If you need to confirm whether a client should use the root domain or /v1, check API Endpoints first.

​
One-Click Setup

curl -fsSL https://raw.githubusercontent.com/xujfcn/crazyrouter-hermes/main/setup.sh | bash
The script prompts for your Crazyrouter API key, lets you choose a default model, and can optionally test the connection.
The setup script writes local Hermes config. Back up ~/.hermes/.env and ~/.hermes/config.yaml first if you already have custom settings.

​
Manual Setup

If you prefer not to run the script, write the Hermes config manually.

​
Environment Variables

Edit ~/.hermes/.env: 
OPENAI_API_KEY=sk-your-crazyrouter-key
OPENAI_BASE_URL=https://api.crazyrouter.com/v1

​
Default Model

Edit ~/.hermes/config.yaml: 
model:
  provider: "custom"
  default: "claude-opus-4-8"
  base_url: "https://api.crazyrouter.com/v1"
Change default to any model allowed by your Crazyrouter token.

​
Switch Models

After setup, switch models inside Hermes with /model: 
/model claude-opus-4-8
/model gpt-5.5
/model claude-opus-4-8
/model gemini-3.1-pro
/model deepseek-chat
/model qwen-max
/model gpt-4o
If switching returns model not allowed or 403, the current Crazyrouter token usually does not allow that model.

​
Verification Checklist

  • OPENAI_API_KEY in ~/.hermes/.env is a valid Crazyrouter token
  • OPENAI_BASE_URL is https://api.crazyrouter.com/v1
  • provider in ~/.hermes/config.yaml is custom
  • the default model is allowed by the Crazyrouter token
  • Hermes can return a test response
  • /model gpt-5.5 or another allowed model switches successfully

​
Common Issues

SymptomLikely causeFix
401 unauthorizedAPI key is wrong, expired, or rotatedcreate a new Crazyrouter token and update ~/.hermes/.env
403 or model not allowedthe token does not allow the selected modelallow that model in the Crazyrouter token settings
Hermes still uses the old providerHermes did not reload config, or the wrong config path was editedrestart Hermes and check ~/.hermes/.env plus ~/.hermes/config.yaml
Connection timeoutlocal network cannot reach Crazyroutercheck proxy, firewall, and DNS settings

View the crazyrouter-hermes repository

Review the setup scripts, manual config example, and latest notes.