Skip to content
@seal-rg

Safety-and-Efficiency Aligned Learning Group

This is the github page of the SEAL research group at the ELLIS Institute Tübingen and the Max Planck Institute for Intelligent Systems

Popular repositories Loading

  1. recurrent-pretraining recurrent-pretraining Public

    Pretraining and inference code for a large-scale depth-recurrent language model

    Python 968 87

  2. streaming streaming Public

    Code for the paper Multi-Stream LLMs: Unblocking Language Models with Parallel Streams of Thoughts, Inputs and Outputs

    Python 70 5

  3. binflash binflash Public

    Triton-based Implementation of Attention supporting partially filled binary masks

    Python 3

Repositories

Showing 3 of 3 repositories
  • streaming Public

    Code for the paper Multi-Stream LLMs: Unblocking Language Models with Parallel Streams of Thoughts, Inputs and Outputs

    seal-rg/streaming's past year of commit activity
    Python 70 5 2 0 Updated Jun 23, 2026
  • binflash Public

    Triton-based Implementation of Attention supporting partially filled binary masks

    seal-rg/binflash's past year of commit activity
    Python 3 MIT 0 0 0 Updated May 20, 2026
  • recurrent-pretraining Public

    Pretraining and inference code for a large-scale depth-recurrent language model

    seal-rg/recurrent-pretraining's past year of commit activity
    Python 968 Apache-2.0 87 1 1 Updated Dec 29, 2025

Top languages

Loading…

Most used topics

Loading…