Once you've set up your columns or chosen a template, the next step is to add files as sources of data for your agents, enabling them to extract, analyze, and generate information effectively.
Uploading files
There are two primary methods for uploading documents to your AI Grid Agents:
1. Row by row
You can upload files individually for each row.
Click the `+` button on each row to open a modal displaying your internal data sources. From here, you can search and select the files that best suit your needs.
You can add as many files as you like. However, keep in mind that the broader the context, the higher the chance Lampi might use semantically related but inconsistent information. Always choose the most coherent and relevant files.
2. Bulk upload
You can upload multiple files at once using the "Add Documents" button at the top of the screen. This bulk upload option distributes the documents row by row, which is particularly useful if you are comparing, or analyzing a large number of documents. You still have the possibility to add other documents to complete the files.
Note: If you haven't selected the Files tool, (but another tool, such as Web Search) you can't add files (except to modify the agent).
Running the AI Grid Agent
Once your documents are uploaded, it's time to run your AI Grid Agents.
To run it, simply select the rows you want the agent to process and click the "Run" button that appears.
The AI agent will begin extracting relevant information from the uploaded documents.
You can monitor the progress of each task, which will display either "Running" (in progress) or "Queued" (waiting to start).
For more granular control, you can also launch specific cells by simply clicking the run button on each cell.
For more, you can refer to 🏁 Running AI Grid Agents.