Skip to main content

​
1. What should I do if my balance hasn’t arrived after topping up?

Normally, top-ups are credited instantly. If there is a delay, please contact support with the following information:
  • Account ID (UUID)
  • Top-up order number (product_xxx)
  • Payment screenshot
Note: The system may occasionally experience delays, and balances are typically credited manually after investigation. If confirmed to be a system issue, the platform usually issues credit vouchers as compensation.

​
2. What is the top-up rate? How do I top up in RMB?

Top-ups are calculated in USD. You can top up directly via Alipay. The real-time exchange rate is displayed on the top-up page.

​
3. How can I avoid top-up failures?

Top-up failures are usually caused by two main reasons:
  • Card issuer declined. This may be due to the following reasons. Please check or contact your card issuer for details:
    • The corresponding payment channel is not activated.
    • The credit card has expired or is frozen.
    • Insufficient credit card balance.
    • Incorrect card number.
    • Incorrect security code.
  • Risk control measures of the payment channel. Please check and make necessary adjustments:
    • Too many cards associated with the device ID.
    • A very high number of declined cards using this email address.
    • Very short time since this card first appeared with this device ID on the Stripe network.
    • Very low authorization rate associated with this email address.
    • The name on the email address does not match the name on the card.

​
4. How is pricing calculated?

Different models/capabilities are charged differently:
  • LLMs / Embedding / Reranker: Billed per token, including input / output / cache write / cache read dimensions.
  • Images / Videos / Audio: Billed per API call.

​
5. How can I view detailed bills and usage?

​
6. Why was I charged a large amount with very few API calls?

Please check the following:
  • Model selection: High-end models like Claude Opus 4.5 consume credits quickly (110K input tokens can cost $0.45 or more)
  • Token calculation: Models like Gemini calculate image input tokens based on resolution; high-resolution images can generate a large number of tokens
  • Cache billing: Some models (e.g., Claude) charge at the “cache input” price when Prompt Caching misses, which may be more expensive than regular input

​
7. Why does it show insufficient balance (403) when I still have vouchers/balance?

Possible reasons include:
  • The account truly has insufficient funds (total of cash + vouchers is not enough)
  • System billing display delay (already fully consumed)
  • Per-minute RPM/TPM rate limit exceeded (error message may display as insufficient balance)

​
8. Why am I being charged for free models?

Please check if you switched models during use. If you switch from a paid model to a free model within a conversation, charges may be incurred due to context token calculations. It is recommended to check the model call records in billing details.

​
9. Are refunds supported? How do I apply?

Generally, used balance is not refundable. If a refund is truly needed (e.g., the platform cannot meet your business requirements), please provide:
  • Account ID (UUID)
  • Registration email
  • Reason for refund
Note: Compatibility issues encountered through third-party tools (e.g., Cursor) are typically not eligible for refunds. It is recommended to test with a small top-up first.

​
10. Are refunds returned to the original payment method?

Yes, refunds are returned to the original payment account. If the top-up was made by a company, please confirm the original payment channel with your finance department.
Contact Support If the above FAQ does not resolve your issue, please contact technical support through the following channels:
  • Email: support@myrouter.ai
  • Please provide the following information:
    • Problem description + screenshots
    • Account ID (UUID)
    • Trace ID (if available, usually found in the error message)
    • Request parameters (with sensitive data redacted)