Connect an agent

OpenClaw

OpenClaw is an open-source AI assistant gateway and bot framework with OpenAI-compatible API support. Point its model provider to Yomi API to use all chat models in the platform catalog.

Prerequisites

Follow the official OpenClaw documentation to download, install and initialize it.
Sign in to the Yomi API console, create a key in API Keys and copy it. It is shown only at creation, so keep it safe.
Ensure your account has enough credit. If needed, add credit through the User Center's Add Credit page.

Configure Yomi API

OpenClaw manages model providers through configuration files. Enter these OpenAI-compatible parameters in its model configuration (the provider section):

API address (base_url): https://api.yomiapi.com/v1
API Key (api_key): YOUR_API_KEY (replace with your key)
Model (model): Use a model ID from the Model Catalog, such as claude-sonnet-5 or gpt-5.6-sol

Example below. Consult the official OpenClaw documentation for the exact field names and configuration file location:

base_url: https://api.yomiapi.com/v1
api_key: YOUR_API_KEY
model: claude-sonnet-5

Use the model ID from the catalog, not its display name. Find the full catalog under Models in the console, or query the /v1/models endpoint.

Start and verify

Save the configuration and restart OpenClaw. Send a test message; a model response confirms successful setup. You can also see the corresponding call under Usage in the console User Center.

If OpenClaw reports an error, first verify the key and address directly with cURL. This helps distinguish platform configuration issues from OpenClaw configuration issues:

curl https://api.yomiapi.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "deepseek-v4-flash",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

Common questions

401 authentication failure

The API key is incorrect or has been deleted. Check it in API Keys; if necessary, create a new key and update your OpenClaw configuration.

404 response or no response

This is usually an address issue. Ensure base_url is https://api.yomiapi.com/v1. Omitting or duplicating /v1 produces an incorrect request path.

Model does not exist or is unavailable

Use the model ID, not the display name. Also check that your key's group includes the model; the Model Catalog lists availability for each group.