Batch Mode enables users to upload a CSV file and run an AI agent on each row of the spreadsheet. Each row is treated as a separate user query, allowing for efficient processing of multiple requests.

Batch Mode processes each request independently, including tool use and invoking AI teammates, just as it would in the ChatUI.

​
Use Cases

  • Mass data analysis
  • Bulk web scraping
  • Large-scale content generation
  • Automated data enrichment

Batch Mode works exceptionally well with the web scrape tool for mass scraping operations.

​
How to Use Batch Mode

Follow these steps to utilize Batch Mode:

1

Access Batch Mode

Navigate to the top right corner of the desired agent and click on the “Batch Mode” button.

2

Upload CSV

Upload the CSV file you want to process.
3

Configure Advanced Options (Optional)

In advanced options, you can:

  • Select specific columns for the agent to consider
  • Choose the starting and ending rows for processing
4

Enter Instructions

Provide the instruction that will be applied to every row of the spreadsheet.

5

Run the Batch

Execute the batch process.
6

Download Results

Download the results in either PDF or CSV format: - CSV: The agent’s output will appear as a new column called “Invicta Output” at the end of the spreadsheet. - PDF: Each output will be on a separate page.

​
Important Considerations

Be mindful of your Compute Points consumption. Batch Mode can quickly use a significant amount of resources, especially for large datasets.

Exercise caution when using tools that may trigger account blocks or hit rate limits. For example, mass email sending could be flagged as spam, and some integrations may have request limitations.

​
Best Practices

  1. Start with a small subset of your data to test the process before running a full batch.
  2. Monitor your Compute Points usage to avoid unexpected depletion.
  3. Exercise caution when using tools that interact with external systems or services. Be mindful of web scraping practices and ensure compliance with target website terms of service. When sending emails, consider spam regulations and recipient preferences. For API calls, respect rate limits and usage policies of third-party services.
  4. Regularly check the output quality to ensure the AI agent is performing as expected across all rows.

When using tools that interact with external systems, always consider the potential impact on those systems and your account status with various services.

By leveraging Batch Mode, you can significantly increase the efficiency and scale of your AI agent operations, enabling you to process large volumes of data with ease.

LLM SettingsCloning