Frequently Asked Questions

If your key is lost, you need to:

1. Delete the old key (optional, recommended for security)
2. Create a new key
3. Update your application or configuration file with the new key

When creating a key, you can:

• Set expiration time: Limit the validity period of the key
• Quota limit: Set the maximum usage quota for this key
• Model permissions: Select which models this key can access

Cubence supports creating multiple API keys. It's recommended to create independent keys for different projects or environments (development, testing, production) for easier management and security control.

After logging into the console, visit the "Wallet" or "Recharge" page to recharge online. Cubence supports multiple payment methods.

Cubence uses pay-as-you-go billing and does not offer monthly subscriptions. Recharge your balance as needed and review the current rates.

After logging into the console, go to the "Billing" or "Consumption Records" page to view detailed API call records and expense details.

Share your referral link from the console. Eligibility, rates and settlement depend on the applicable program and order; review the current rules and your reward records in the console.

Please troubleshoot with these steps:

1. Check if API Token is correct

   • Confirm Token is copied correctly without extra spaces
   • Confirm Token has not expired

2. Check if Base URL is correct

   • Confirm the endpoint matches your selected route; see Endpoints
   • Claude Code and Gemini CLI use the root URL; OpenAI-compatible clients such as Codex require /v1

3. Check account balance

   • Confirm account has sufficient balance
   • Check balance status in the console

4. Check rate limits

   • Check error messages to confirm if rate limit is exceeded
   • Reduce request frequency or contact support to increase limits

Use the protocol and Base URL documented for your selected client and model, together with a Cubence API key. Compatibility depends on the endpoint and supported features. Follow the relevant setup guide rather than assuming every upstream API feature is supported.

See the public model catalog for active model identifiers and the pricing page for base rates. Confirm group access and regional availability in Model Plaza before use.

After logging into the console, go to the "Logs" or "Call Records" page to view:

• Request time
• Model used
• Token consumption
• Response status
• Error messages (if any)

Please follow the configuration tutorial for the corresponding tool. Supported tools include: Claude Code, Codex, Gemini CLI.

Common configuration issues:

1. API Token not set correctly: Check configuration files or environment variables
2. Base URL error: Confirm the selected route; Claude Code and Gemini CLI use the root URL, while Codex requires /v1
3. Wrong configuration file path: Check if the configuration file is in the correct location
4. Permission issues: Ensure configuration files have correct read/write permissions

Each device needs to configure CLI tools separately:

1. Create API Token in Cubence console
2. Follow the configuration tutorial on the new device
3. You can use the same Token or create independent Tokens for different devices

Try these methods:

• Access via mobile device
• Force refresh on PC:
  • Windows: Ctrl + F5 or Ctrl + Shift + R
  • Mac: Cmd + Shift + R
• Mobile: Long press refresh button and select "Force Refresh"
• Clear browser cache and revisit

Use the pre-sales Telegram group for questions before purchasing. For account, order or API issues, use the after-sales Telegram group entry or QQ group listed under pre-sales and after-sales support.

Check published announcements on the Cubence website and the relevant setup documentation.

If the above content doesn't solve your problem, choose a channel under pre-sales and after-sales support, or check other documentation for more information.