Hello Google AI Team and Community,
I am a web developer building an intelligent document archiving system for a university in Iraq using Laravel/PHP. We are currently testing the Gemini API on the free tier and would like guidance on moving to a paid production environment.
The application will process images of Arabic official correspondence, extract their text and metadata, and store the validated results alongside the original images in our database.
Our expected usage is:
-
Approximately 100 registered users.
-
Approximately 100 document images per day in total across all users, not per user.
-
Approximately 3,000 images per month.
-
Each image is between 800 KB and 5 MB.
-
Some users may submit documents at the same time.
-
The primary language is Arabic, including document reference numbers and dates.
We need the API to return structured JSON containing:
-
title: the document subject or a brief subject summary. -
book_no: the document’s own reference number, distinguished from numbers of other letters mentioned in the text. -
book_date: the document’s issue