Usage and Billing

Yunxi AI calculates usage based on model, input and output tokens, key group, cache behavior, and pricing rules. You can inspect request logs and balance changes in the console.

What you can see

• Current balance or quota.
• Usage by API key.
• Request model, tokens, cost, latency, and error details.
• Aggregated usage by model and key.

Add Balance with a Redemption Card Code

Yunxi AI currently uses redemption card codes to add balance. The purchase page is hosted by Liandong Shop:

https://pay.ldxp.cn/shop/6IKDTWVQ

Recommended flow:

1. Open the Yunxi AI console and go to Wallet.
2. Click Open Liandong Shop.
3. Choose a Yunxi points product in Liandong Shop and complete payment.
4. Copy the delivered card code.
5. Return to the Yunxi AI Wallet page, paste the card code into Redeem card code, and click Redeem.

If the shop page opens but no card code is delivered, use the order query or contact entry on Liandong Shop first, then contact Yunxi AI support if the code still cannot be confirmed.

Investigating unexpected cost

Check:

• The model name used by the request.
• The group attached to the API key.
• Input and output tokens.
• Cache hit or miss details.
• Billing details in request logs.
• Whether a higher-capability or longer-context model was used.

If the cost still looks wrong, contact support with request time and model name.