Tab | Managed Object |
Models | Various large models supported by the platform (language, vision, multimodal understanding, vector, speech models), managing the pay-as-you-go billing for their free trial packages and default model services. |
Plugins | Various plugin capabilities provided by the platform, managing their activation status and pay-as-you-go billing. |
Column | Description | |
Model Name | | Click a model name to go to the corresponding default inference service details. If the free package has not been claimed for the model, the default inference service endpoint has not been created, and the jump cannot be performed. |
Model Rate Limiting | | Displays the TPM / RPM quota of the model. |
Model Status | Online | Indicates that the model is online and running normally. |
| Pending offline | Indicates that the model has a published offline plan. The model can still be called normally before the offline date. Migrate to the recommended alternative model as soon as possible. |
| Offline | Indicates that the model has stopped service and can no longer be called. Call requests will return errors. |
| End of marketing | Indicates that the model no longer supports activation or purchase by new users. Existing users who have already activated the model can continue to use it until the model is offline. |
Column | Description |
Claim Status | Unclaimed: You can click the button to claim the free trial package for this model. Claimed: The button is grayed out and cannot be used to claim again. Not Supported: This model does not support free trials. To use it, enable pay-as-you-go billing. |
Free quota balance | Displays the remaining free quota for claimed models in the form of remaining percentage + progress bar. Hover over the progress bar to view details: free quota tokens, consumed tokens, remaining tokens, and remaining percentage %. This field is not displayed for models that have not claimed or do not support free quotas. Supports header filtering by "Nearly exhausted (<20%)" to filter models with less than 20% free quota remaining, so you can enable pay-as-you-go billing in time to avoid service interruptions. |
Free quota expiration time | Displays the expiration time of the free quota. This field is not displayed for models that have not claimed or do not support free quotas. Supports header filtering by "Expiring soon" to filter models with less than 7 days until expiration, so you can enable pay-as-you-go billing in time to avoid service interruptions. |
Column | Description |
Activation Status | Not Enabled: You can click the button to enable pay-as-you-go billing for this model. Enabled: You can click the button to disable pay-as-you-go billing for this model. A secondary confirmation dialog box pops up for both enabling and disabling. |
Billing Mode | Currently supports Token-based billing and points-based billing. For models without pay-as-you-go billing enabled, this column does not display information. |
Page | Responsibilities |
Activation Management | Model-level overview entry for viewing model lists, claiming free quotas, and enabling/disabling pay-as-you-go billing, helping users perform lightweight activation. |
Online Inference Service | Service-level management entry. A model can have multiple inference services (1 default service with the same name + n custom services). You can create custom services, configure rate limiting and billing, and make API calls as needed. |
피드백