Skip to content

@swiss-ai swiss-ai

Change the repository type filter

All

    Repositories list

    14 repositories

    • nanotron

      Public
      Minimalistic large language model 3D-parallelism training
      Python
      •
      Apache License 2.0
      •122•7•1•5•Updated Nov 10, 2024Nov 10, 2024

    • llm-pretrain-data-toxicity-removal

      Public
      Python
      •
      Apache License 2.0
      •0•0•0•0•Updated Nov 6, 2024Nov 6, 2024

    • nanotron-multilingual

      Public
      A copy of nanotron for multilingual training
      Python
      •
      Apache License 2.0
      •122•0•0•2•Updated Oct 23, 2024Oct 23, 2024

    • lighteval-multilingual

      Public
      LightEval is a lightweight LLM evaluation suite that Hugging Face has been using internally with the recently released LLM data processing library datatrove and LLM training library nanotron.
      Python
      •
      MIT License
      •95•2•0•0•Updated Oct 18, 2024Oct 18, 2024

    • video2dataset

      Public
      Easily create large video dataset from video urls
      Python
      •
      MIT License
      •65•1•0•1•Updated Oct 14, 2024Oct 14, 2024

    • llm-proxy

      Public
      LLM Serving and User Control
      Python
      •1•0•0•0•Updated Aug 30, 2024Aug 30, 2024

    • containers

      Public
      Containers for multimodal initiative (and maybe more across Swiss AI?)
      Dockerfile
      •0•0•0•0•Updated Aug 7, 2024Aug 7, 2024

    • ml-4m-v2

      Public
      0•0•0•0•Updated Aug 5, 2024Aug 5, 2024

    • ml-4m

      Public
      4M: Massively Multimodal Masked Modeling (NeurIPS 2023 Spotlight)
      Python
      •
      Apache License 2.0
      •95•0•13•4•Updated Aug 5, 2024Aug 5, 2024

    • vllm

      Public
      A high-throughput and memory-efficient inference and serving engine for LLMs
      Python
      •
      Apache License 2.0
      •4.6k•0•0•0•Updated Jul 31, 2024Jul 31, 2024

    • data-tooling

      Public
      Tool set for data preparation and selection in the context of Swiss-AI (forked from DataTrove)
      Python
      •
      Apache License 2.0
      •147•0•0•0•Updated Jul 15, 2024Jul 15, 2024

    • data-PDF-pipeline

      Public
      PDF pipeline for creating training corpora (mainly for llm, multimodal and alignment horizontals)
      Python
      •
      Apache License 2.0
      •0•0•0•0•Updated May 8, 2024May 8, 2024

    • MoE

      Public
      some mixture of experts architecture implementations
      Python
      •
      Apache License 2.0
      •1•7•0•0•Updated Mar 22, 2024Mar 22, 2024

    • Megatron-LLM

      Public
      distributed trainer for LLMs
      Python
      •
      Other
      •77•0•0•0•Updated Feb 8, 2024Feb 8, 2024