Which file types and sizes does InstantDrive accept?

Last updated: 5 November 2026

What is the size limit?

Each file can be up to 100 MB. A larger file is refused with an error that states the 100 MB limit, both in the web app and over the WebDAV drive. The limit applies per file, not per upload batch. Split very large scans or compress video before it reaches InstantDrive, because the product is meant for business documents and not for raw video.

Which file types does InstantDrive read?

The upload picker does not filter by type, so you can select any file. What happens next depends on the extension.

Other formats, such as Word and Excel files, are not on the reading list. They are stored without fields, so add the category and dates yourself in Edit Metadata.

What is compressed or converted?

PDFs are re-optimised, and the new version is kept only if it is smaller than the original. JPEG, PNG, BMP, TIFF and WebP images are converted to WebP at quality 80, and any image longer than 2,500 pixels on its longest side is scaled down to that size. The converted file is kept only if it is smaller. Otherwise the original bytes are stored.

Everything else is stored unchanged. Because scans end up as WebP, a stored file does not always end in .pdf.

What about HEIC and AVIF photos?

HEIC and AVIF files are accepted but not converted. They are stored as uploaded. Whether text can be read from them is not verified, so a phone photo in one of these formats may arrive with no fields. Export the photo as JPEG or scan it to PDF before uploading, or use the phone camera setting that saves JPEG.

How are scans and text PDFs treated differently?

A PDF with a text layer is read directly. For a long PDF, the first five pages are read, plus the last page when the file has more than five pages. A PDF with no text layer is treated as a scan. Its first page is rendered and read with OCR.

Images go straight to OCR. If the OCR result is weak, nothing was read, or the pattern engine could not recognise the type, the image is sent to Google Gemini as a fallback. Passwords and damaged files are handled gracefully. A password-protected or corrupted PDF is stored with a generic category and its name taken from the file name, and no fields are read.

How are duplicates handled?

On web upload, InstantDrive computes a SHA-256 hash of the file. If an active file with the same hash already exists in your company, the upload shows as verified in the vault and no second copy is stored. A file you deleted no longer counts.

Two different files that end up with the same name in one folder are both kept. The second gets a number suffix such as _1. Uploads over the WebDAV drive are matched by path, not by hash, so the same content copied twice under different names is stored twice.

How are files named after upload?

The name is generated from what was read, in the pattern Prefix_Entity_Identifier_Exp followed by the expiry date, for example a trade licence or an Emirates ID. If no expiry date exists, the issue date, an amount or the upload date is used. The target convention across InstantDrive documentation is DocCategory_EntityName_Identifier_ValidityDate, and the real output follows it closely but not exactly.

What happens as storage fills up?

Each company has a recorded storage limit, set to 2 TB when the workspace is created. InstantDrive keeps a usage figure that increases by the stored size of each upload and decreases when a document is deleted. The upload code does not check the usage figure against the limit before accepting a file. Plan storage amounts are listed on the Pricing page.

Deleting a document marks it as deleted and updates the usage figure. The underlying object is not removed, and the app has no trash view or restore.

What should you do next?

Convert HEIC and AVIF photos to JPEG, remove password protection from PDFs, and split anything near 100 MB. Then upload a test batch and check the result in Edit Metadata. Questions about limits go to [email protected].