Skip to content

@swiss-ai swiss-ai

Change the repository type filter

All

    Repositories list

    66 repositories

    • Megatron-LM

      Public
      Ongoing research training transformer models at scale
      Python
      •
      Other
      •3.6k•43•6•18•Updated Feb 7, 2026Feb 7, 2026

    • benchmark-image-tokenzier

      Public
      Jupyter Notebook
      •2•0•0•1•Updated Feb 7, 2026Feb 7, 2026

    • multimodal-data

      Public
      Jupyter Notebook
      •0•0•0•1•Updated Feb 5, 2026Feb 5, 2026

    • gh200-wheels

      Public
      Python wheels and images for GH200 GPUs
      Dockerfile
      •
      Apache License 2.0
      •0•0•0•0•Updated Feb 4, 2026Feb 4, 2026

    • tokenizer-intrinsic-evals

      Public
      A suite of intrinsic evaluation metrics for the Apertus tokenization team to use during tokenizer development
      Python
      •8•1•0•0•Updated Feb 2, 2026Feb 2, 2026

    • benchmark-audio-tokenizer

      Public
      Python
      •1•0•0•1•Updated Feb 2, 2026Feb 2, 2026

    • perf-check

      Public
      Perf-Check is a lightweight “canary” suite to verify the AI training stack before large runs. It quickly checks GPU compute, HBM bandwidth, NVLink/PCIe P2P, NCC…
      Cuda
      •0•1•0•0•Updated Jan 30, 2026Jan 30, 2026

    • sglang

      Public
      SGLang is a fast serving framework for large language models and vision language models.
      Python
      •
      Apache License 2.0
      •4.4k•0•0•0•Updated Jan 29, 2026Jan 29, 2026

    • verl

      Public
      verl: Volcano Engine Reinforcement Learning for LLMs
      Python
      •
      Apache License 2.0
      •3.2k•0•0•0•Updated Jan 28, 2026Jan 28, 2026

    • nanotron_climllama

      Public
      Minimalistic large language model 3D-parallelism training
      Python
      •
      Apache License 2.0
      •0•1•0•0•Updated Jan 28, 2026Jan 28, 2026

    • model-spinning

      Public archive
      Python
      •2•9•0•0•Updated Jan 27, 2026Jan 27, 2026

    • Swiss-AI-Romansh-Scripts

      Public
      The set of scripts was developed to process Rumantsch data for Apertus V1, the LLM created by the Swiss AI Initiative.
      Python
      •
      Apache License 2.0
      •1•4•0•0•Updated Jan 26, 2026Jan 26, 2026

    • mmore

      Public
      Massive Multimodal Open RAG & Extraction A scalable multimodal pipeline for processing, indexing, and querying multimodal documents Ever needed to take 8000 P…
      Python
      •
      Apache License 2.0
      •37•186•10•10•Updated Jan 23, 2026Jan 23, 2026

    • lmms-eval

      Public
      One-for-All Multimodal Evaluation Toolkit Across Text, Image, Video, and Audio Tasks
      Python
      •
      Other
      •512•0•0•1•Updated Jan 21, 2026Jan 21, 2026

    • ml4science-Image-Audio-Text-Instruction-Data

      Public
      Jupyter Notebook
      •0•0•0•0•Updated Jan 14, 2026Jan 14, 2026

    • project2

      Public
      Jupyter Notebook
      •1•0•0•2•Updated Jan 14, 2026Jan 14, 2026

    • fm-service

      Public
      newer llm service
      Python
      •3•1•0•0•Updated Jan 13, 2026Jan 13, 2026

    • apertus-probes

      Public
      Code and notebooks for DSL #16 hallucination probe project
      Jupyter Notebook
      •0•2•0•0•Updated Jan 10, 2026Jan 10, 2026

    • posttraining-data

      Public
      Python
      •0•4•1•0•Updated Dec 28, 2025Dec 28, 2025

    • lsaie-amd-project

      Public
      Final project for the course "Large-Scale AI Engineering". Porting Megatron-LM-based models to AMD GPU hardware.
      Python
      •1•0•0•0•Updated Dec 19, 2025Dec 19, 2025

    • ml4science-apertus-rag-evaluation

      Public
      Python
      •0•0•0•0•Updated Dec 18, 2025Dec 18, 2025

    • evals

      Public
      Python
      •3•6•0•1•Updated Dec 17, 2025Dec 17, 2025

    • reasoning_getting-started

      Public
      Shell
      •0•3•0•0•Updated Dec 16, 2025Dec 16, 2025

    • parity-aware-bpe

      Public
      Parity-Aware Byte-Pair Encoding: Improving Cross-lingual Fairness in Tokenization [arXiv 2025]
      tokenizationbpemultilingual-nlpllmsmultilingual-tokenization
      Python
      •
      MIT License
      •3•18•1•0•Updated Dec 10, 2025Dec 10, 2025

    • apertus-format

      Public
      Response format to be used with apertus
      Python
      •1•11•0•0•Updated Dec 3, 2025Dec 3, 2025

    • torrent

      Public
      Python
      •1•1•0•0•Updated Nov 27, 2025Nov 27, 2025

    • Pai-Megatron-Patch

      Public
      The official repo of Pai-Megatron-Patch for LLM & VLM large scale training developed by Alibaba Cloud.
      Python
      •
      Apache License 2.0
      •225•0•0•1•Updated Nov 4, 2025Nov 4, 2025

    • pretrain-data

      Public
      Pretraining data reconstruction scripts for Apertus
      Python
      •
      Apache License 2.0
      •10•113•2•1•Updated Oct 27, 2025Oct 27, 2025

    • apertus-finetuning-recipes

      Public
      Python
      •13•25•1•1•Updated Oct 22, 2025Oct 22, 2025

    • lm-evaluation-harness

      Public
      A framework for few-shot evaluation of language models.
      Python
      •
      MIT License
      •3k•7•0•2•Updated Oct 22, 2025Oct 22, 2025