Start routing AI agent requests through the gateway in four steps.
Make requests through the gateway
See the example request below
export API_KEY="YOUR_API_KEY"curl --request POST \
--url https://platform.weirwood.ai/api/v1/responses \
--header 'authorization: Bearer $API_KEY' \
--header 'content-type: application/json' \
--data '{"model": "gpt-5.4-nano","input": [{"role": "user","content": "1-2=?"}]}'Configure the gateway as a custom model provider in OpenClaw.
Add the following configuration to ~/.openclaw/openclaw.json, replacing YOUR_API_KEY with your key from the API keys page.
{
"agents": {
"defaults": {
"workspace": "/home/alan/.openclaw/workspace",
"models": {
"gateway/gpt-5.4-nano": {}
},
"model": {
"primary": "gateway/gpt-5.4-nano"
}
}
},
"models": {
"providers": {
"gateway": {
"baseUrl": "https://platform.weirwood.ai/api/v1",
"apiKey": "YOUR_API_KEY",
"api": "openai-responses",
"models": [
{
"id": "gpt-5.4-nano",
"name": "gateway/gpt-5.4-nano"
}
]
}
}
}
}Configure the gateway as a custom model provider in Codex.
Add the following to ~/.codex/config.toml, then export your key from the API keys page as YOUR_API_KEY.
model_provider = "gateway"
[model_providers.gateway]
name = "Gateway"
base_url = "https://platform.weirwood.ai/api/v1"
env_key = "YOUR_API_KEY"Configure the gateway as an model gateway in Claude Code.
Add the following to ~/.claude/settings.json, replacing YOUR_API_KEY with your key from the API keys page.
{
"env": {
"ANTHROPIC_BASE_URL": "https://platform.weirwood.ai/api",
"ANTHROPIC_AUTH_TOKEN": "YOUR_API_KEY"
}
}