Introduction to AI Models Hub
The AI Models Hub in Docsumo is a powerful resource designed to simplify the process of document data extraction. With a collection of over 50 pre-trained document types spanning across various categories, you can effortlessly extract data from your documents right from the first upload.

Here's everything you need to know about the AI Models Hub:
Pre-Trained Document Types
The AI Models Hub houses more than 50 pre-trained document types, each tailored to specific categories.
These categories include Income, Procurement, Energy, Identification, Insurance, Tax Forms documents.
No initial training is required, you can start extracting data immediately.
- When using pre-trained document types, you'll find that the majority of fields are readily available during the document review process.
- Review and validate your document data with ease, benefiting from the predefined extraction rules.
- Extracted data is accurate and relevant, saving you valuable time and effort.
Adding Specific Fields:
While using pre-trained document types, if you require specific fields that aren't included by default, we offer a simple solution to add these specific fields to the extraction list, tailoring the process to your needs.
This flexibility allows you to fine-tune the extraction to match your document's unique structure.
Customisation
You can confidently rely on the pre-trained document types to handle common document structures.
For specialised or unique documents, the option to create custom document types ensures no document is left behind.
- If the available pre-trained document types don't precisely match your requirements, Docsumo provides a user-friendly option to create custom document types.
- Simply upload samples of the document type you need and train a model on top of them.
- This customisation feature ensures that you can handle even the most unique document types efficiently.
Updated over 1 year ago