TinyDiT

TinyDiT is an 85 million parameter unconditional image generation model trained on 21,000+ anime face images.

Model Details

• Model Name: TinyDiT
• Architecture: Diffusion Transformer (DiT-inspired)
• Parameters: 85M
• Task: Unconditional Image Generation
• Dataset Size: 21,000+ anime face images
• VAE: Lightweight 13M parameter VAE
• Generation Type: Anime face generation from random noise (no text conditioning)
• Image Size: 64x64px
• Github Repo: https://github.com/Nitesh1405/TinyDiT/tree/main

Dataset

TinyDiT was trained on a curated anime face dataset containing over 21k images.

Dataset Repository: huggan/anime-faces

VAE

The model uses a compact 13M parameter Variational Autoencoder (VAE) for latent-space encoding and decoding.

Example Generated Images

Below is a sample images generated by TinyDiT:

Usage

• HuggingFace Space: https://huggingface.co/spaces/nitesh501/TinyDiT

git clone https://github.com/Nitesh1405/TinyDiT.git && cd TinyDiT

pip install -r requirements.txt

python app.py
#the model will automatically download on first run if you have wget, if not you can download the model from https://huggingface.co/nitesh501/tinydit and place it in TinyDit Folder.

Limitations

• Trained only on anime face data
• Unconditional generation only
• Limited diversity compared to larger diffusion models
• May occasionally generate blurry or distorted outputs

Acknowledgements

Inspired by DiT architectures, latent diffusion models, and the open-source generative AI community.