Model HQ

Batch Run

Learn how to process multiple documents simultaneously with Model HQ's Batch Run feature

 

Batch Run Overview

In this Section, we will explore using the Batch Run capabilities of Model HQ.

Batch Run is designed to automate the execution of an Agent process across multiple documents at once, without the user having to load each document one-by-one.

batch run

We will use the Contract Analyzer example we covered in Agents to provide an example of a Batch Run and how to use this feature. (A video of this agent process running in Batch Run mode can also be found in our YouTube Channel: https://youtu.be/FSjpAgIZnPM)

 

Uploading Documents for Batch Processing

Once you have selected 'Batch Run', you will be directed to the Upload Documents for Batch Processing page.

batch run

 

From here, select for our example the first 5 agreements from the sample files that are provided in the Model HQ package that can be found under c:\users\{user name}\llmware_data\sample_files\agreements

batch run

 

Then select '>' to start the batch run.

 

Batch Run Results

Within a few minutes, each of the 5 contracts will be processed in this Batch Run. Here are the examples of the results:

batch run

 

Once the Batch Run is completed, the user will have access to the results via a Word document or a JSON file, depending on the use case.

Batch Run can be applied to almost any Agent workflow where a user needs to apply the workflow against a number of various files.


Conclusion

The Batch Run feature significantly enhances productivity by allowing you to process multiple documents in a single operation. This is particularly valuable for tasks like contract analysis, document classification, or data extraction across large document sets.

If you have any questions or feedback about the Batch Run feature, please reach out to us at support@aibloks.com.