OpenClaw

OpenClaw custom providers are configured through models.providers. For rout.my, use the OpenAI-compatible openai-completions adapter and the /v1 base URL.

Save your key

export ROUTMY_API_KEY="sk_your_key_here"

Ready model IDs

This config is already filled with rout.my defaults:

Role	Model
Primary	openai/gpt-5.4
Fallback	openai/gpt-5.4-mini

You only need to change them if you intentionally want another exact ID from /v1/models.

Example config

Save this to ~/.openclaw/openclaw.json.

{
  "agents": {
    "defaults": {
      "model": {
        "primary": "rout/openai/gpt-5.4",
        "fallbacks": ["rout/openai/gpt-5.4-mini"]
      },
      "models": {
        "rout/openai/gpt-5.4": { "alias": "Primary via rout.my" },
        "rout/openai/gpt-5.4-mini": { "alias": "Fallback via rout.my" }
      }
    }
  },
  "models": {
    "mode": "merge",
    "providers": {
      "rout": {
        "api": "openai-completions",
        "baseUrl": "https://api.rout.my/v1",
        "apiKey": "${ROUTMY_API_KEY}",
        "models": [
          { "id": "openai/gpt-5.4", "name": "openai/gpt-5.4" },
          { "id": "openai/gpt-5.4-mini", "name": "openai/gpt-5.4-mini" }
        ]
      }
    }
  }
}

OpenClaw uses rout/ as the provider prefix and sends the remaining model ID to rout.my.

Validate

openclaw config validate
openclaw gateway start

Test the key

curl https://api.rout.my/v1/chat/completions \
  -H "Authorization: Bearer $ROUTMY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-5.4",
    "messages": [{ "role": "user", "content": "Hello!" }]
  }'