Skip to content
@mit-han-lab

MIT HAN Lab

Efficient AI Computing. PI: Song Han

Pinned Loading

  1. streaming-llm streaming-llm Public

    [ICLR 2024] Efficient Streaming Language Models with Attention Sinks

    Python 6.8k 379

  2. llm-awq llm-awq Public

    [MLSys 2024 Best Paper Award] AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration

    Python 2.9k 239

  3. efficientvit efficientvit Public

    Efficient vision foundation models for high-resolution generation and perception.

    Python 2.7k 216

  4. bevfusion bevfusion Public archive

    [ICRA'23] BEVFusion: Multi-Task Multi-Sensor Fusion with Unified Bird's-Eye View Representation

    Python 2.5k 451

  5. temporal-shift-module temporal-shift-module Public

    [ICCV 2019] TSM: Temporal Shift Module for Efficient Video Understanding

    Python 2.1k 419

  6. once-for-all once-for-all Public

    [ICLR 2020] Once for All: Train One Network and Specialize it for Efficient Deployment

    Python 1.9k 341

Repositories

Showing 10 of 60 repositories
  • x-attention Public

    XAttention: Block Sparse Attention with Antidiagonal Scoring

    mit-han-lab/x-attention’s past year of commit activity
    Python 108 2 0 0 Updated Mar 26, 2025
  • llm-awq Public

    [MLSys 2024 Best Paper Award] AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration

    mit-han-lab/llm-awq’s past year of commit activity
    Python 2,879 MIT 239 153 8 Updated Mar 26, 2025
  • nunchaku Public

    [ICLR2025 Spotlight] SVDQuant: Absorbing Outliers by Low-Rank Components for 4-Bit Diffusion Models

    mit-han-lab/nunchaku’s past year of commit activity
    Cuda 1,039 Apache-2.0 68 92 (5 issues need help) 1 Updated Mar 25, 2025
  • mit-han-lab/ComfyUI-nunchaku’s past year of commit activity
    Python 77 Apache-2.0 1 8 0 Updated Mar 20, 2025
  • omniserve Public

    [MLSys'25] QServe: W4A8KV4 Quantization and System Co-design for Efficient LLM Serving; [MLSys'25] LServe: Efficient Long-sequence LLM Serving with Unified Sparse Attention

    mit-han-lab/omniserve’s past year of commit activity
    C++ 616 Apache-2.0 41 37 5 Updated Mar 6, 2025
  • torchsparse Public

    [MICRO'23, MLSys'22] TorchSparse: Efficient Training and Inference Framework for Sparse Convolution on GPUs.

    mit-han-lab/torchsparse’s past year of commit activity
    Cuda 1,291 MIT 150 33 2 Updated Feb 24, 2025
  • torchquantum Public

    A PyTorch-based framework for Quantum Classical Simulation, Quantum Machine Learning, Quantum Neural Networks, Parameterized Quantum Circuits with support for easy deployments on real quantum computers.

    mit-han-lab/torchquantum’s past year of commit activity
    Jupyter Notebook 1,427 MIT 213 61 (4 issues need help) 9 Updated Feb 21, 2025
  • VisCompare Public

    A WebUI for Side-by-Side Comparison of Media (Images/Videos) Across Multiple Folders

    mit-han-lab/VisCompare’s past year of commit activity
    Python 21 Apache-2.0 2 0 0 Updated Feb 21, 2025
  • deepcompressor Public

    Model Compression Toolbox for Large Language Models and Diffusion Models

    mit-han-lab/deepcompressor’s past year of commit activity
    Python 392 Apache-2.0 32 41 1 Updated Feb 21, 2025
  • Block-Sparse-Attention Public

    A sparse attention kernel supporting mix sparse patterns

    mit-han-lab/Block-Sparse-Attention’s past year of commit activity
    C++ 169 BSD-3-Clause 6 6 0 Updated Feb 13, 2025