AI Models lists the AI models configured in the platform, organized by provider. From this section, you can create, update, and delete models.
You can access AI Models under AI Settings in the Console Menu. This section is visible only to users with the System administrator role.
The list displays each model's Model Name, Type, Priority, Status, Context Window, and whether a cost is defined (Cost Defined?). You can filter the list by Provider, Model name, and Status using the fields at the top right of the page. To export the list, click Export and select Excel.
Each model has two available actions: Update to edit its configuration, and Delete to remove it.
To create a new model, click Create New and complete the following fields:
- Id: Generated automatically by the platform.
- Provider: Provider this model belongs to.
- Model: Model identifier.
- Description (JSON): JSON-formatted description of the model.
- Is Custom: Indicates whether the model is custom.
- Type: Model type (for example, Chat).
- Output Tokens: Maximum number of output tokens.
- Context Window: Size of the model context window.
- Status: Indicates whether the model is Active, Hidden, or Deprecated.
- Priority: Numeric value that determines the order in which models are used. Lower values indicate higher priority.
- Parameters (JSON): Additional model parameters in JSON format.
Click Confirm to save the configuration, or Cancel to discard the changes.
Clicking a model name opens the model detail view, which contains three tabs:
- General
- Deployments
- AIModel Cost
Displays the model configuration fields: Id, Provider, Model, Description (JSON), Is Custom, Type, Output Tokens, Context Window, Status, Priority, and Parameters (JSON). Use Update to edit the configuration or Delete to remove the model.
Lists the provider deployments associated with the model. The list displays the Provider, Provider Deployment Name, Provider Deployment Description, and Provider Deployment Status. Use Create New to associate a new deployment with the model.
Lists the cost revisions defined for this model. The list displays the Cost Revision, whether it is active (Active?), Creation Timestamp, and Cost default JSON. Use Create New to add a cost revision, or Update and Delete to manage existing ones.
The cost is defined in JSON format using the following keys:
- prompt_tokens: Cost per consumed prompt token.
- completion_tokens: Cost per generated completion token.
- total_tokens: Total token cost.
Some models support differentiated cost schemes based on token type, such as cached tokens within prompt tokens. For pricing details, see the provider's official documentation: