Pinned Loading

1. vid2cleantxt vid2cleantxt Public

   Python API & command-line tool to easily transcribe speech-based video files into clean text

   Jupyter Notebook 228 29

2. textsum textsum Public

   CLI & Python API to easily summarize text-based files with transformers

   Python 132 10

3. megalodon-jax megalodon-jax Public

   jax rewrite for CEMA support + other training speedups.

   Python

4. NeoBERT NeoBERT Public

   fork of NeoBERT refactored for easy pretraining & arch experiments

   Python 2

5. decoder-pytorch-template decoder-pytorch-template Public template

   Hackable PyTorch template for decoder-only transformer architecture experiments. Llama baseline with RoPE, SwiGLU, RMSNorm. Swap components, train, compare

   Python 2

6. parakit parakit Public

   daemon style local dictation kit powered by Parakeet-TDT & CrispASR

   Rust 3 1