Instructions to use Johnny-Z/Anima-Light-Lavender with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusion Single File
How to use Johnny-Z/Anima-Light-Lavender with Diffusion Single File:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
Clarification on underscores in tags
#2
by Espamholding - opened
Normally, tags (like character names) are trained with spaces and not underscores.
In the anima.json workflow, underscores are used for the character's name mari_(idol)_(blue_archive) and they persist in the output of the custom node. The custom node also retains spaces if they are used for character/artist/franchise name, etc.
Is this a mistake, or was the model trained with underscores? Does this apply to artist and franchise names and extra tags?
Character names follow the same convention as Danbooru: no title-casing, with spaces replaced by underscores. That said, for an LLM-based text encoder such as Qwen3-0.6B, minor deviations in formatting are unlikely to matter much.