# IA Monitoring

"**AI Monitoring**" centralizes the detailed tracking of resource consumption and the performance of your artificial intelligence models, agents, and flows.

To access this area, simply click on the "**AI Monitoring**" menu in the left sidebar.

<figure><img src="/files/1jcJZ994YiDJsBFZguYh" alt=""><figcaption></figcaption></figure>

### Plan and Consumption

This section presents data regarding your plan and token usage, where you can view the volume of tokens consumed during the selected period relative to the total contracted limit.

To view the available models, click on the plan name.

### Overview

The Overview provides a temporal analysis of AI resource usage through charts and tables.

* **Date Filter:** Data can be filtered by period using the calendar selector in the upper right corner.
* **Export:** To download the data, select the desired period, choose the format (CSV or JSON), and click "Export history".

### Charts

Below are the types of charts presented in "**AI Monitoring**".

* **Agent tokens used:** Volume of tokens consumed by agent interactions during the period.
* **Agent executions:** Total number of registered agent triggers.
* **Model tokens used:** Volume of tokens consumed directly by the language models.
* **Model executions:** Total number of calls made to the models.
* **Workflow tokens used:** Total tokens consumed within structured workflows.
* **Workflow executions:** Total number of workflow executions completed during the period.

#### Execution Details

The following tables are available to detail executions within the selected period:

**List of executions by Agent**

Provides the volume of executions for each agent created in the system:

* **Agent Name:** Identification of the monitored agent.
* **Tokens used:** Volume consumed specifically by this agent.
* **Total executions:** Number of times the agent processed a task.

**List of executions by Model**

Allows for auditing the demand sent to each AI provider:

* **Model ID:** Unique identifier of the model.
* **Model Name:** Commercial name of the model (e.g., OpenAI, Llama).
* **Tokens used:** Total accumulated consumption by the model during the selected period.
* **Total executions:** Volume of executions performed, with a sorting option (highest/lowest) via the column filter.

**List of executions by Workflow**

Allows for analyzing the efficiency of automations performed during the period:

* **Workflow Name:** Identification of the automation (e.g., "AI AGENT FLOW").
* **Tokens used:** Token cost generated by the full execution of the flow.
* **Total executions:** Frequency with which the workflow was triggered.

***

### FAQ - Frequently asked questions about AI Monitoring

<details>

<summary>What can be tracked in the "AI Monitoring" area?</summary>

It centralizes the detailed monitoring of resource consumption and the performance of AI models, agents, and workflows.

</details>

<details>

<summary>How can I filter consumption data by a specific period?</summary>

Use the calendar selector located in the upper right corner of the "Overview" section.

</details>

<details>

<summary>Is it possible to export the resource usage history?</summary>

Yes, select the desired period, choose the format (CSV or JSON), and click "Export history".

</details>

<details>

<summary>What metrics do the Overview charts present?</summary>

They show the volume of tokens used and the total number of executions for agents, models, and workflows.

</details>

<details>

<summary>Where can I find consumption details by AI provider (e.g., OpenAI)?</summary>

In the "List of executions by Model" table, which identifies the model's commercial name, tokens used, and total executions.

</details>


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.skyone.cloud/english/skyone-studio/artificial-intelligence/ia-monitoring.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.