Join the conversation

Join the community of Machine Learners and AI enthusiasts.

Sign Up
mkurman 
posted an update 5 days ago
Post
3609
Introducing a new architecture, MedIT One – a single-token transformer with LSTM-like recurrence.

It is extremely fast in training and inference, but we lack funding for large-scale training. Enjoy 🍓

https://github.com/MedITSolutionsKurman/medit-one

In this post