Photo Namer

Rename photos using short AI-generated descriptions from a local LM Studio server.

Requirements

• Python 3.9+
• LM Studio running a local vision-capable model
• Local server started at http://localhost:1234

Setup

1. Create and activate a virtual environment.
2. Install dependencies:

pip install -r requirements.txt

3. Start LM Studio:
   • Open LM Studio
   • Load a vision-capable model
   • Local Server tab -> Start Server

Usage

Dry-run (default):

python rename_photos.py "C:\\path\\to\\photos"

Apply renames:

python rename_photos.py "C:\\path\\to\\photos" --apply

Recursive scan:

python rename_photos.py "C:\\path\\to\\photos" --recursive --apply

How It Works

• Each image is sent to the local LM Studio server for a 3-6 word description.
• The description is slugified into a filename.
• Collisions are avoided by adding a numeric suffix.
• By default, the tool prints a plan without renaming.

Supported Formats

jpg, jpeg, png, webp, bmp, gif

Notes

• If the server is not reachable, the script exits with a clear message.
• Filenames are capped at 60 characters.