Skip to content
Change the repository type filter

All

    Repositories list

    • MegaDLMs

      Public
      GPU-optimized framework for training diffusion language models at any scale. The backend of Quokka, Super Data Learners, and OpenMoE 2 training.
      Python
      33000Updated Apr 23, 2026Apr 23, 2026
    • Merge of megatron-train, autoexperiment and oellm_pretrain.
      Python
      2890Updated Apr 23, 2026Apr 23, 2026
    • JudgeArena

      Public
      Evaluating LLM with swappable judges: local, remote, openrouter on multiple benchmarks.
      Python
      Apache License 2.0
      41054Updated Apr 23, 2026Apr 23, 2026
    • Ongoing research training transformer models at scale
      Python
      Other
      3.9k000Updated Apr 22, 2026Apr 22, 2026
    • training-data-catalogue

      Public
      Curated Public Repository of LLM (Pre-)Training Data
      Shell
      26390Updated Apr 21, 2026Apr 21, 2026
    • oellm-cli

      Public
      Python
      81261Updated Apr 21, 2026Apr 21, 2026
    • Python
      1000Updated Apr 20, 2026Apr 20, 2026
    • Report slurm compute usage on Discord automatically every week.
      Shell
      Apache License 2.0
      0010Updated Apr 20, 2026Apr 20, 2026
    • Repo for post-training LLMs
      Python
      3411Updated Apr 17, 2026Apr 17, 2026
    • Datamix model scripts for LUMI
      Shell
      MIT License
      0000Updated Apr 17, 2026Apr 17, 2026
    • About Utility scripts for converting models with Megatron-Bridge
      Jinja
      0010Updated Apr 16, 2026Apr 16, 2026
    • Python
      Apache License 2.0
      0100Updated Apr 8, 2026Apr 8, 2026
    • Python
      Apache License 2.0
      0100Updated Apr 7, 2026Apr 7, 2026
    • Python
      1000Updated Mar 26, 2026Mar 26, 2026
    • Python
      0000Updated Mar 6, 2026Mar 6, 2026
    • notebooks

      Public
      Jupyter Notebook
      Apache License 2.0
      0100Updated Mar 4, 2026Mar 4, 2026
    • Allow to patch opensci models to run them with recent transformers versions
      Python
      Apache License 2.0
      0000Updated Feb 24, 2026Feb 24, 2026
    • simple test A vs B models
      Python
      0200Updated Feb 23, 2026Feb 23, 2026
    • Setup environment variables and slurm configuration automatically on EuroHPC clusters
      Shell
      0100Updated Jan 29, 2026Jan 29, 2026
    • Ongoing research training transformer models at scale
      Python
      Other
      3.9k000Updated Jan 27, 2026Jan 27, 2026
    • Python
      1200Updated Jan 22, 2026Jan 22, 2026
    • Shell
      0000Updated Jan 7, 2026Jan 7, 2026
    • MegaTron open-sci fork
      Python
      Other
      3.9k000Updated Oct 14, 2025Oct 14, 2025
    • Python
      0000Updated Oct 2, 2025Oct 2, 2025
    • Evaluate a list of models and tasks
      Python
      Other
      2010Updated Aug 18, 2025Aug 18, 2025
    • Python
      Apache License 2.0
      0000Updated Jul 29, 2025Jul 29, 2025
    • MultiSynt

      Public
      MultiSynt: an open multilingual synthetic dataset for LLM pre-training.
      0010Updated Jun 2, 2025Jun 2, 2025
    • Taskboard

      Public
      Apache License 2.0
      011210Updated Apr 14, 2025Apr 14, 2025
    ProTip! When viewing an organization's repositories, you can use the props. filter to filter by custom property.