Transform how you organize and analyze your visual assets with AI Autotagger, the ultimate plugin to harness the power of the latest AI models. Automatically tag, describe, and analyze images and videos in your library with unprecedented detail and accuracy. Beyond just photography, AI Autotagger can recognize UI components, brands/companies, devices, and much more, ensuring comprehensive categorization for all creative assets.
Advanced AI-Driven Tagging
Leverage the intelligence of the multitude of AI models available to automatically generate detailed tags and descriptions for your images and videos. Whether it's identifying lighting and color theory, artistic styles, UI components, or even specific brands and devices, AI Autotagger ensures your assets are meticulously categorized and easily searchable.
Customizable Analysis Rules with Presets
Tailor the plugin to meet your specific needs using customizable analysis rules organized into presets. AI Autotagger comes with Photography, UI Design, and Fashion Inspiration presets by default, allowing you to quickly start analyzing and categorizing your assets based on the context that matters most to you. Also included are presets to rename your items based on their content, or extract text from your items.
Bring Your Own Key (BYOK) Model
AI Autotagger operates on a "Bring Your Own Key" approach. You'll need an account from one of the providers (OpenAI, Google, or Anthropic) with an API key to get started, OR you can use local inference with providers like LM Studio and Ollama. This setup ensures that you only pay for what you use, directly through the provider you choose, giving you full control over your usage and costs.
Fixed: LM Studio and Ollama base URLs couldn't be changed
Fixed: maxTokens error
Added option to process items using their thumbnail images (e.g., SVGs)
Added support for tag modifiers with configurable position and spacing
Improved processing for very large inputs/outputs (e.g., long documents)
Version 7 (Included in this release)
Added Ollama and LM Studio support for local model inference
Improved model selection UI, including allowing custom model names so you can now use the latest models as soon as they're released
Added option to resize items before AI processing (improves speed/cost)
Added Gemini Flash 8B model
Improved UI freezing issues when processing many items
Added detailed logs for each item
Fixed: Item names longer than 245 characters are now truncated to avoid data corruption
Added Ko-Fi button for those who want to support the project
Auto-refresh items when you change the selection in the main Eagle window
Copy and paste presets — share presets with people!
Option to send low-res versions of images/videos to reduce API cost
Improve reliability of Google's Gemini 1.5 models
Various bug fixes and UI improvements
Fix double-paste API key bug on Windows