EDICO: Open-Source App to caption and curate a consistent character dataset for LoRA training
Hello there!
Since 2018, I've been creating small scripts for different Generative AI Tools and more. Now, I've merged them all and created some applications with them; one of the finished ones is EDICO.
📝 EDICO is a local, browser-based tool for building character LoRA training datasets with AI-generated captions, and it turns a folder of character images into a clean, captioned training dataset — no command-line juggling, no manual caption typing.
✨ The app is absolutely free, Open-Source, with an MIT license, so you can try it, modify it, etc. It's written in PowerShell and pulls Ollama models.
How does it work? The 7-step workflow:
👋 Welcome: This screen is the first step, and here you create your project or make a new one.
👁️ Vision Model: You select your favorite Vision Model to create the captioning. It's required to have Ollama installed.
🪪 Identity: In this step, you choose your trigger word and also the basic and static identity for your character.
🖼️ Images: You drag and drop images or an entire folder, obviously with the images of your character.
📁 Categories: You can select, add, or remove categories that your model will use for captioning your character images.
⚡ Pipeline: This step is the process where your vision model will start creating the captions of your images.
📂 Organizer: Here you will organize your captioned images, edit the captions, and do a lot more; this area here is a masterpiece, trust me!
For the entire documentation and user guides, follow the following links:
About EDICO: https://gabrielx.com/edico-eliz-delias-image-captioner-and-organizer/
User Guide: https://gabrielx.com/docs-edico-documentation-user-guide-dataset-recommendations/
GitHub Repo: https://github.com/gabrielxcom/edico
More free AI Tools: https://gabrielx.com/open-source-mit-ai-tools/
¿Hablas español? La documentación también está disponible en español:
🔗 https://gabrielx.com/es/herramientas-ia-de-codigo-abierto-licencia-mit/
Support us!
You can support us by listening to our songs (yes, as silly as that). Visit: www.IamElizAi.com 🎶