adi-mish/miniformer
Miniformer is a lightweight PyTorch transformer library for researchers, educators, and tinkerers. It supports encoder-only and encoder–decoder models with multi-head attention and rotary embeddings. The codebase is small and readable, ideal for prototyping or edge deployment.
No commits in the last 6 months.
Stars
—
Forks
—
Language
Python
License
GPL-3.0
Category
Last pushed
Aug 03, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/adi-mish/miniformer"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
philipperemy/keras-attention
Keras Attention Layer (Luong and Bahdanau scores).
tatp22/linformer-pytorch
My take on a practical implementation of Linformer for Pytorch.
lucidrains/fast-weight-attention
Implementation of Fast Weight Attention
datalogue/keras-attention
Visualizing RNNs using the attention mechanism
ematvey/hierarchical-attention-networks
Document classification with Hierarchical Attention Networks in TensorFlow. WARNING: project is...