Skip to main content

How does cache works in GLM models?

The caching mechanism is currently undergoing open beta testing. Therefore, the underlying logic of the caching mechanism (including but not limited to: how cache hits are triggered, how long cached content is retained, etc.) has not yet been announced. A part of your request content may be cached on our cloud platform, and the storage of cached content is currently completely free. During the request process, if your request triggers a cache hit, the cached content will be calculated at approximately 1/5 of the original price.

Why hasn’t my account balance changed after I used the API?

  • Please be advised that there is currently a processing delay in our billing system. You do not need to worry too much if you see your account balance kept the same after using api port.
  • The billing history reflect daily consumption records, and therefore display the billing status from the previous day (n-1).
  • Current day consumption will not be immediately visible in the billing details.

How can I check the rate limits?

You can check the rate limits from here: https://z.ai/manage-apikey/rate-limits

How should I recharge?

You can recharge from the billing page: https://z.ai/manage-apikey/billing

Why can’t I recharge when using credit card?

When using a credit card to recharge, please ensure that you are not using 3DS verification. 3DS verification is not supported in our platform at this moment.