Skip to main content
For first-time use, run the setup wizard to complete all configuration: 
openclaw onboard

​
Step 1: Select Model Provider

In the Model/auth provider list, scroll to the bottom and select: 
Custom Provider (Any OpenAI or Anthropic compatible endpoint)

​
Step 2: Enter API Base URL

In the API Base URL field, enter the APIYI endpoint: 
https://api.apiyi.com/v1

​
Step 3: Choose API Key Input Method

When asked how to provide the API key, select: 
Paste API key now

​
Step 4: Paste API Key

Paste your APIYI token (also called API key, starting with sk-) into the input field. How to get your token:
  1. Open the APIYI Token Management page: https://api.apiyi.com/token
  2. Find the default token with type Pay-per-use Priority
  3. On the far right of that token’s row, under the “Actions” column, click the first copy button
  4. This copies the sk- prefixed token (API key) — paste it here
Paste API Key
No need to create a new token — just use the default “Pay-per-use Priority” token generated by the system.

​
Step 5: Select Endpoint Compatibility

In the Endpoint compatibility option, select: 
OpenAI-compatible (Uses /chat/completions)
If you primarily use Claude models and need native features like Prompt Caching, you can select Anthropic-compatible instead. See Anthropic Native Configuration for details.

​
Step 6: Set Model ID

In the Model ID field, enter the model name you want to use, for example: 
gpt-5.4
Other options: claude-sonnet-4-6, deepseek-v3.2, gemini-3.1-pro-preview, etc.

​
Step 7: Verify and Complete

OpenClaw will automatically verify the configuration. Upon success, the system generates an Endpoint ID (e.g., custom-api-apiyi-com), indicating the setup is complete.
The verification process may occasionally return errors — retrying usually succeeds. If it keeps failing, check that your API key is correct and your network connection is stable.

​
Configuration Summary

ParameterValue
ProviderCustom Provider
API Base URLhttps://api.apiyi.com
CompatibilityOpenAI-compatible
Endpoint IDcustom-api-apiyi-com (auto-generated)

​
Modify Configuration

After initial setup, use the configure command to re-enter interactive configuration: 
openclaw configure
This opens an interactive configuration menu for:
  • Model providers and default model
  • Chat channel settings
  • Skill enable/disable
  • Gateway parameters

​
Single Config Commands

Quickly view or modify individual config items: 
{/* View current config value */}
openclaw config get agents.defaults.model.primary

{/* Set config value */}
openclaw config set agents.defaults.model.primary "apiyi/gpt-5.4"

{/* Set API key */}
openclaw config set models.providers.apiyi.apiKey "sk-your-key"

​
Web UI Config Panel

After starting the Dashboard, you can also manage configuration in the browser: 
openclaw dashboard
Open http://127.0.0.1:18789/ to access the settings page where you can:
  • Visually edit model configuration
  • Manage chat channel connections
  • View and toggle configured skills
  • Monitor logs and running status in real-time

​
Start the Service

After configuration, start the Gateway: 
openclaw gateway start
After configuration changes, restart the service:
openclaw gateway restart