Relm API + MCP.

Uploads & indexing

Per-unit document uploads

Attach leases, floor plans, or work orders to a specific unit so chat can cite them when answering unit-level questions.

Relm TeamUpdated 2 min read
On this page

Beyond property-level uploads (rent roll, OM, P&L), Relm also supports per-unit uploads — attaching documents to a specific unit so chat and AI Summary can use them in unit-level analysis.

What's a per-unit upload

Examples:

  • Unit 312's lease and renewal notices.
  • A floor plan for unit type A1.
  • A lease snippet confirming a Section 8 contract on a specific unit.
  • A maintenance work-order log for a problem unit.

These attach to the unit, not to the property as a whole.

How to upload

  1. Open the property and click the Unit Details card to open the Units page.
  2. Click the unit to open its detail view.
  3. In the unit's Documents section, click Upload Document.
  4. Choose one or more files.

You can also tag a file to a unit while uploading it the usual way: in the Upload Documents window, open the unit picker (Assign to Unit) and choose the unit instead of Entire Property.

The file is OCR'd and indexed scoped to that unit.

Photos: to add photos to the unit's gallery, use Upload Images in the unit's detail view (JPEG, PNG, or WebP). Chat reads only the text in a document, so a photo uploaded as a document won't tell it anything about the unit's condition.

What it enables

  • Chat queries can cite the unit-specific upload when answering. Asking "When does 312's lease end?" returns an answer grounded in the lease, naming the file it came from.
  • Deep Search can use unit documents when assessing renovation status and estimated market rent for that unit.
  • AI Summary can reference flagged anomalies on specific units ("Unit 312 has documented water damage per attached work order").

Limits

  • File size: up to 50 MB per file.
  • Format: same as property-level uploads (PDF, image, DOCX, XLSX, etc.).

Privacy

Per-unit uploads are scoped to the same organization-level access as everything else in Relm. They're encrypted at rest and never used to train models. See Privacy & data handling.

What's next

Was this article helpful?

Still need help?

Our team usually responds within one business day. Tell us what you're trying to do and we'll get you unstuck.

Relm Pro Help Center