Skip to content
@swiss-ai

swiss-ai

Popular repositories Loading

  1. mmore mmore Public

    Massive Multimodal Open RAG & Extraction A scalable multimodal pipeline for processing, indexing, and querying multimodal documents Ever needed to take 8000 PDFs, 2000 videos, and 500 spreadsheets …

    Python 186 37

  2. apertus-tech-report apertus-tech-report Public

    Tech Report of the Apertus LLM Suite

    129 4

  3. pretrain-data pretrain-data Public

    Pretraining data reconstruction scripts for Apertus

    Python 113 10

  4. Megatron-LM Megatron-LM Public

    Forked from NVIDIA/Megatron-LM

    Ongoing research training transformer models at scale

    Python 43 20

  5. MoE MoE Public

    some mixture of experts architecture implementations

    Python 25 3

  6. apertus-finetuning-recipes apertus-finetuning-recipes Public

    Python 25 13

Repositories

Showing 10 of 66 repositories
  • Megatron-LM Public Forked from NVIDIA/Megatron-LM

    Ongoing research training transformer models at scale

    swiss-ai/Megatron-LM’s past year of commit activity
    Python 43 3,646 6 18 Updated Feb 7, 2026
  • swiss-ai/benchmark-image-tokenzier’s past year of commit activity
    Jupyter Notebook 0 2 0 1 Updated Feb 7, 2026
  • swiss-ai/multimodal-data’s past year of commit activity
    Jupyter Notebook 0 0 0 1 Updated Feb 5, 2026
  • gh200-wheels Public

    Python wheels and images for GH200 GPUs

    swiss-ai/gh200-wheels’s past year of commit activity
    Dockerfile 0 Apache-2.0 0 0 0 Updated Feb 4, 2026
  • tokenizer-intrinsic-evals Public Forked from cimeister/tokenizer-analysis-suite

    A suite of intrinsic evaluation metrics for the Apertus tokenization team to use during tokenizer development

    swiss-ai/tokenizer-intrinsic-evals’s past year of commit activity
    Python 1 8 0 0 Updated Feb 2, 2026
  • swiss-ai/benchmark-audio-tokenizer’s past year of commit activity
    Python 0 1 0 1 Updated Feb 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, NCCL collectives, and MPI latency/bandwidth, with optional filesystem tests. It outputs structured pass/fail summaries and flags regressions versus a baseline.

    swiss-ai/perf-check’s past year of commit activity
    Cuda 1 0 0 0 Updated Jan 30, 2026
  • sglang Public Forked from sgl-project/sglang

    SGLang is a fast serving framework for large language models and vision language models.

    swiss-ai/sglang’s past year of commit activity
    Python 0 Apache-2.0 4,365 0 0 Updated Jan 29, 2026
  • verl Public Forked from verl-project/verl

    verl: Volcano Engine Reinforcement Learning for LLMs

    swiss-ai/verl’s past year of commit activity
    Python 0 Apache-2.0 3,218 0 0 Updated Jan 28, 2026
  • nanotron_climllama Public

    Minimalistic large language model 3D-parallelism training

    swiss-ai/nanotron_climllama’s past year of commit activity
    Python 1 Apache-2.0 0 0 0 Updated Jan 28, 2026