VisionTagger

Overview

VisionTagger uses local AI vision models to automatically generate titles, descriptions, keywords, and structured metadata for your photos in bulk. All processing happens entirely on your Mac with no files uploaded and no per-image fees. The app ships with six preconfigured vision models including Qwen3-VL, Gemma 3, InternVL3, and Pixtral, and supports custom GGUF-compatible models with tunable parameters like temperature, top-P, and top-K. You can enrich results with context hints, GPS location data, and camera metadata, then export in multiple formats including XMP sidecars, JSON, CSV, TXT, and Finder tags. ExifTool integration enables embedded metadata writing. VisionTagger supports JPEG, PNG, TIFF, HEIC, WebP, and RAW formats, includes automatic translation to 18+ languages, and integrates with Apple Shortcuts for automation workflows.

Pricing: €24.99 (launch offer) | Free trial: 100 images

Minimum macOS: 26.0 (Tahoe)

Architecture: Apple Silicon

Key Features

• Batch metadata generation with customizable titles, descriptions, keywords, and structured fields
• Six preconfigured vision models including Qwen3-VL 8B/30B, Gemma 3 4B, InternVL3 8B, and Pixtral 12B
• Support for custom GGUF-compatible models with tunable parameters
• Context hints, GPS location, and camera metadata integration for richer results
• Export to XMP sidecars, JSON, CSV, TXT, and Finder tags
• ExifTool integration for embedded metadata writing
• Supports JPEG, PNG, TIFF, HEIC, WebP, and RAW formats
• Automatic translation to 18+ languages
• Apple Shortcuts integration for automation
• Real-time generation preview
• All processing happens locally on your Mac — no files uploaded

Tags

Similar Apps