Dashboard

The Dashboard page provides a comprehensive and detailed view of artificial intelligence services usage and costs on the platform. It is an essential tool for monitoring and managing resource consumption. Reports in the dashboard are displayed by month. You can navigate to previous months using the month selector in the top right corner of the page.

Costs

The Costs section presents charts showing the total accumulated costs throughout the month. These charts help identify consumption trends and plan future resource usage.

• Bar Chart: Displays daily accumulated costs throughout the month, allowing detailed analysis of when and how resources are consumed.
• Pie Chart: Shows the accumulated cost distribution throughout the entire month, providing an overview of each service's contribution to total costs.

💡 Chart Customization

You can customize the charts to show specific information:

• Click on service names (like Text to Speech, Speech to Text, etc.) above the charts to add or remove their respective information.
• Colors help differentiate each type of cost, facilitating personalized visualization.

Usage by Model

This section details the usage of different AI models available on the platform. Each model has individual charts showing:

• API Requests: Number of API calls made to the model.
• Tokens/Characters/Seconds: Amount of resources consumed by the model, varying according to the type of service.

Note

Usage by model charts have different scales for the Y-axis. This is done to ensure usage variations are clearly visible, but prevents direct comparisons between different models.

Usage by Instance

Here you can visualize the usage of rented machine instances to run specific models. Charts in this section show:

• Duration: Total instance usage time.
• Cost: Accumulated cost from instance usage.

Note

Like usage by model charts, usage by instance charts also have different scales for the Y-axis, ensuring clarity of usage and cost variations, but preventing direct comparisons between different instances.

Detailed Usage Table

The table at the bottom of the dashboard provides a complete breakdown of all recent activities on the platform, including:

• Date and Time: When the activity occurred.
• Model: Which AI model was used.
• Type: Type of service executed (e.g., Embedding, Speech to Text, Text to Speech, Text Generation).
• Execution Type: Specific details of the execution type.
• Quantity: Amount of resources consumed (e.g., tokens, seconds).
• Cost: Cost associated with each activity.

Switching between Tables

You can switch between viewing the usage table by model and by instance by clicking on the respective tabs (Model and Instance).

Data Export

You can export usage data for external analysis:

1. Access Export: Click the Export button in the top right corner of the detailed usage table.
2. Select Export Options:
   • Table Type: Choose between Model or Instance.
   • Format: Choose the file format, CSV or JSON.
   • Period: Select the desired period for export (maximum 180 days).

Filters and Search

The detailed usage table has filter and search options to facilitate finding specific information:

• Filter by Model: Allows filtering activities by AI model.
• Filter by Type: Allows filtering activities by service type.
• Filter by Execution Type: Allows filtering activities by execution type.
• Filter by Cost: Allows filtering activities by cost range.

Using the Dashboard

The dashboard is an essential tool for managing resource usage and controlling costs on the platform. Use the charts to monitor daily and monthly usage, identify models that are consuming more resources, and adjust your usage strategies as needed to optimize costs and efficiency.