Skip to content

@eth-easl EASL

Change the repository type filter

All

    Repositories list

    33 repositories

    • modyn

      Public
      Modyn is a research-platform for training ML models on growing datasets.
      Python
      •
      MIT License
      •6•43•95•7•Updated Feb 18, 2025Feb 18, 2025

    • deltazip-ae

      Public
      Artifact Evaluation for DeltaZip
      Jupyter Notebook
      •
      Apache License 2.0
      •0•0•0•0•Updated Feb 14, 2025Feb 14, 2025

    • deltazip

      Public
      Compression for Foundation Models
      Jupyter Notebook
      •
      Apache License 2.0
      •3•24•0•1•Updated Feb 14, 2025Feb 14, 2025

    • cloud-comp-arch-project

      Public
      Starter code for semester project in Cloud Computing Architecture course at ETH Zurich
      Python
      •8•4•0•0•Updated Feb 13, 2025Feb 13, 2025

    • gpu-util-interference

      Public
      CUDA benchmarks for measuring GPU utilization and interference
      Cuda
      •
      MIT License
      •1•4•0•0•Updated Feb 11, 2025Feb 11, 2025

    • torchtitan

      Public
      A native PyTorch Library for large model training
      Python
      •
      BSD 3-Clause "New" or "Revised" License
      •276•0•0•0•Updated Feb 7, 2025Feb 7, 2025

    • mixtera-model-eval

      Public
      Nuts and bolts for evaluation of models trained in context of mixtera
      Python
      •0•0•0•0•Updated Feb 7, 2025Feb 7, 2025

    • torchtitan-mixtera

      Public
      A native PyTorch Library for large model training
      Python
      •
      BSD 3-Clause "New" or "Revised" License
      •276•0•0•1•Updated Feb 5, 2025Feb 5, 2025

    • triteia

      Public
      Useful Kernels for ML in Triton
      Cuda
      •
      Apache License 2.0
      •0•1•2•0•Updated Feb 3, 2025Feb 3, 2025

    • orion

      Public
      An interference-aware scheduler for fine-grained GPU sharing
      Python
      •
      MIT License
      •19•122•9•1•Updated Jan 26, 2025Jan 26, 2025

    • mixtera-clariden

      Public
      temporary nuts and bolts for running mixtera trainings on clariden
      Python
      •0•0•0•0•Updated Jan 24, 2025Jan 24, 2025

    • vidur

      Public
      A large-scale simulation framework for LLM inference
      Python
      •
      MIT License
      •57•0•0•0•Updated Dec 11, 2024Dec 11, 2024

    • dirigent

      Public
      Dirigent: Lightweight Serverless Orchestration
      kubernetesserverlessfaasorchestratorcontainerdfaas-platformfirecracker
      Go
      •
      MIT License
      •3•30•0•0•Updated Dec 8, 2024Dec 8, 2024

    • llm-compressor

      Public
      Transformers-compatible library for applying various compression algorithms to LLMs for optimized deployment with vLLM
      Python
      •
      Apache License 2.0
      •83•0•0•0•Updated Nov 28, 2024Nov 28, 2024

    • pccheck

      Public
      Python
      •
      MIT License
      •0•1•0•0•Updated Nov 8, 2024Nov 8, 2024

    • pecan-experiments

      Public
      Contains instructions and scripts for the ATC'24 Pecan artifact evaluation.
      Python
      •
      Apache License 2.0
      •1•0•0•0•Updated Oct 14, 2024Oct 14, 2024

    • fmengine

      Public
      Utilities for Training Very Large Models
      Python
      •10•57•4•0•Updated Sep 25, 2024Sep 25, 2024

    • cachew

      Public
      ML Input Data Processing as a Service. This repository contains the source code for Cachew (built on top of TensorFlow).
      C++
      •
      Apache License 2.0
      •75k•36•0•2•Updated Sep 10, 2024Sep 10, 2024

    • nanotron

      Public
      Minimalistic large language model 3D-parallelism training
      Python
      •
      Apache License 2.0
      •148•0•0•0•Updated Jul 19, 2024Jul 19, 2024

    • mlibc

      Public
      Portable C standard library
      C
      •
      MIT License
      •140•0•0•0•Updated Jul 15, 2024Jul 15, 2024

    • cglm-metadata

      Public
      Hosts CGLM metadata
      0•0•0•0•Updated Jun 6, 2024Jun 6, 2024

    • serving

      Public
      Kubernetes-based, scale-to-zero, request-driven compute
      Go
      •
      Apache License 2.0
      •1.2k•0•0•0•Updated May 7, 2024May 7, 2024

    • cloudlab_extension

      Public
      Copy node connection information easily
      JavaScript
      •0•1•0•0•Updated Mar 13, 2024Mar 13, 2024

    • rWasm

      Public
      A cross-platform high-performance provably-safe sandboxing Wasm-to-native compiler
      Rust
      •7•0•0•0•Updated Jan 14, 2024Jan 14, 2024

    • cachew_experiments

      Public
      ML Input Data Processing as a Service
      Python
      •
      Apache License 2.0
      •2•8•0•1•Updated Oct 20, 2023Oct 20, 2023

    • elastic-learning-rate-evaluation

      Public
      Python
      •
      Other
      •0•0•0•0•Updated Aug 16, 2023Aug 16, 2023

    • airflow

      Public
      Python
      •
      Apache License 2.0
      •1•0•0•1•Updated Feb 17, 2023Feb 17, 2023

    • varuna

      Public
      Python
      •29•0•0•0•Updated Jul 25, 2022Jul 25, 2022

    • adaptdl

      Public
      Resource-adaptive cluster scheduler for deep learning training.
      Python
      •
      Apache License 2.0
      •79•0•0•0•Updated May 24, 2022May 24, 2022

    • elastic

      Public
      PyTorch elastic training
      Python
      •
      BSD 3-Clause "New" or "Revised" License
      •100•0•0•0•Updated Apr 14, 2022Apr 14, 2022