Skip to content

@IST-DASLab IST Austria Distributed Algorithms and Systems Lab

Change the repository type filter

All

    Repositories list

    50 repositories

    • EvoPress

      Public
      Python
      •1•14•0•0•Updated Dec 17, 2024Dec 17, 2024

    • MicroAdam

      Public
      This repository contains code for the MicroAdam paper.
      Python
      •
      Apache License 2.0
      •4•14•0•0•Updated Dec 14, 2024Dec 14, 2024

    • PanzaMail

      Public
      Python
      •
      Apache License 2.0
      •15•270•4•5•Updated Dec 9, 2024Dec 9, 2024

    • GridSearcher

      Public
      GridSearcher simplifies running grid searches for machine learning projects in Python, emphasizing parallel execution and GPU scheduling without dependencies on SLURM or other workload managers.
      Python
      •
      Apache License 2.0
      •0•2•0•0•Updated Dec 4, 2024Dec 4, 2024

    • llm-foundry

      Public
      LLM training code for Databricks foundation models
      Python
      •
      Apache License 2.0
      •534•0•0•1•Updated Nov 27, 2024Nov 27, 2024

    • marlin_artifact

      Public
      Python
      •0•0•0•0•Updated Nov 25, 2024Nov 25, 2024

    • LDAdam-anonymous

      Public
      0•0•0•0•Updated Nov 20, 2024Nov 20, 2024

    • LDAdam

      Public
      LDAdam - Adaptive Optimization from Low-Dimensional Gradient Statistics
      Python
      •
      Apache License 2.0
      •0•5•0•0•Updated Nov 6, 2024Nov 6, 2024

    • torch_cgx

      Public
      Pytorch distributed backend extension with compression support
      C++
      •
      GNU Affero General Public License v3.0
      •0•16•4•0•Updated Oct 17, 2024Oct 17, 2024

    • ISTA-DASLab-Optimizers

      Public
      Python
      •
      Apache License 2.0
      •0•7•0•0•Updated Sep 5, 2024Sep 5, 2024

    • Sparse-Marlin

      Public
      Boosting 4-bit inference kernels with 2:4 Sparsity
      Cuda
      •
      Apache License 2.0
      •4•61•1•0•Updated Sep 4, 2024Sep 4, 2024

    • marlin

      Public
      FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.
      kernelquantization4bitllm
      Python
      •
      Apache License 2.0
      •52•671•26•4•Updated Sep 4, 2024Sep 4, 2024

    • sparsegpt

      Public
      Code for the ICML 2023 paper "SparseGPT: Massive Language Models Can Be Accurately Pruned in One-Shot".
      Python
      •
      Apache License 2.0
      •98•751•15•1•Updated Aug 20, 2024Aug 20, 2024

    • peft-rosa

      Public
      A fork of the PEFT library, supporting Robust Adaptation (RoSA)
      Python
      •
      Apache License 2.0
      •3•13•1•0•Updated Aug 16, 2024Aug 16, 2024

    • AutoGPTQRoSA

      Public
      Python
      •
      MIT License
      •0•0•0•0•Updated Jun 27, 2024Jun 27, 2024

    • spops

      Public
      C++
      •
      Apache License 2.0
      •0•7•2•0•Updated Jun 20, 2024Jun 20, 2024

    • Mathador-LM

      Public
      Code for the EMNLP 2024 paper "Mathador-LM: A Dynamic Benchmark for Mathematical Reasoning on LLMs".
      Python
      •
      Apache License 2.0
      •0•8•1•0•Updated Jun 18, 2024Jun 18, 2024

    • SPADE

      Public
      Code of SPADE: Sparsity Guided Debugging for Deep Neural Networks
      Jupyter Notebook
      •0•1•0•0•Updated May 25, 2024May 25, 2024

    • QUIK

      Public
      Repository for the QUIK project, enabling the use of 4bit kernels for generative inference - EMNLP 2024
      C++
      •
      Apache License 2.0
      •13•175•5•1•Updated Apr 16, 2024Apr 16, 2024

    • FastOBQ-

      Public
      GPTQ with finetuning
      0•0•0•0•Updated Mar 27, 2024Mar 27, 2024

    • gptq

      Public
      Code for the ICLR 2023 paper "GPTQ: Accurate Post-training Quantization of Generative Pretrained Transformers".
      Python
      •
      Apache License 2.0
      •158•2k•22•1•Updated Mar 27, 2024Mar 27, 2024

    • RoSA

      Public
      Python
      •
      Apache License 2.0
      •3•36•1•0•Updated Feb 13, 2024Feb 13, 2024

    • SparseFinetuning

      Public
      Repository for Sparse Finetuning of LLMs via modified version of the MosaicML llmfoundry
      Python
      •
      Apache License 2.0
      •6•40•3•0•Updated Jan 15, 2024Jan 15, 2024

    • CAP

      Public
      Repository for Correlation Aware Prune (NeurIPS23) source and experimental code
      Python
      •
      Apache License 2.0
      •1•5•1•0•Updated Nov 29, 2023Nov 29, 2023

    • qmoe

      Public
      Code for the paper "QMoE: Practical Sub-1-Bit Compression of Trillion-Parameter Models".
      Python
      •
      Apache License 2.0
      •22•262•3•0•Updated Nov 3, 2023Nov 3, 2023

    • ZipLM

      Public
      Code for the NeurIPS 2023 paper: "ZipLM: Inference-Aware Structured Pruning of Language Models".
      0•2•1•0•Updated Oct 20, 2023Oct 20, 2023

    • TACO4NLP

      Public
      Task aware compression for various NLP tasks
      Python
      •0•2•0•0•Updated Oct 9, 2023Oct 9, 2023

    • KDVR

      Public
      Code for the experiments in Knowledge Distillation Performs Partial Variance Reduction, NeurIPS 2023
      Python
      •
      Apache License 2.0
      •0•1•0•0•Updated Oct 6, 2023Oct 6, 2023

    • sparseprop

      Public
      C++
      •
      Apache License 2.0
      •4•13•0•0•Updated Sep 27, 2023Sep 27, 2023

    • EFCP

      Public
      The repository contains code to reproduce the experiments from our paper Error Feedback Can Accurately Compress Preconditioners available below:
      computer-visiondeep-learningtop-klanguage-modellow-rank-approximationsecond-order-optimizationggtadaptive-gradient-methodfull-matrix-preconditioningm-fac
      Python
      •
      Apache License 2.0
      •0•4•0•0•Updated Sep 12, 2023Sep 12, 2023