Skip to content

@mit-han-lab MIT HAN Lab

All

    Repositories list

    72 repositories

    • kernel-design-agents

      Public
      MIT License
      •84•951•1•1•Updated Sep 2, 2026Sep 2, 2026

    • mlsys2026-flashinfer-contest-solution

      Public
      Python
      •
      MIT License
      •2•32•0•0•Updated Aug 31, 2026Aug 31, 2026

    • KernelWiki

      Public
      Python
      •
      MIT License
      •55•440•1•1•Updated Aug 26, 2026Aug 26, 2026

    • ncu-report-skill

      Public
      Python
      •
      MIT License
      •34•220•0•1•Updated Aug 26, 2026Aug 26, 2026

    • torchquantum

      Public
      A PyTorch-based framework for Quantum Classical Simulation, Quantum Machine Learning, Quantum Neural Networks, Parameterized Quantum Circuits with support for e…
      machine-learningsystemdeep-learning
      machine-learningsystemdeep-learningneural-networkquantumpytorchquantum-computingquantum-machine-learningquantum-simulationml-for-systems
      Jupyter Notebook
      •
      MIT License
      •262•1.7k•70•26•Updated Jul 6, 2026Jul 6, 2026

    • mlsys2026-flashinfer-contest

      Public
      Python
      •5•125•0•0•Updated Jun 13, 2026Jun 13, 2026

    • SMEPO

      Public
      Python
      •1•16•0•0•Updated May 27, 2026May 27, 2026

    • lpd

      Public
      [ICLR 2026 Oral] Locality-aware Parallel Decoding for Efficient Autoregressive Image Generation
      accelerationimagenetimage-generation
      accelerationimagenetimage-generationautoregressiveefficient-algorithmparallel-decoding
      Python
      •
      MIT License
      •7•105•1•0•Updated May 8, 2026May 8, 2026

    • foreact

      Public
      [CVPR 2026 Highlight] ForeAct: Steering Your VLA with Efficient Visual Foresight Planning
      Python
      •6•91•2•0•Updated May 1, 2026May 1, 2026

    • vcpo

      Public
      [ICML 2026] Stable Asynchrony: Variance-Controlled Off-Policy RL for LLMs
      Python
      •
      Apache License 2.0
      •3•34•1•5•Updated Apr 27, 2026Apr 27, 2026

    • vlash

      Public
      Real-Time VLAs via Future-state-aware Asynchronous Inference.
      Python
      •
      Apache License 2.0
      •34•495•18•1•Updated Apr 22, 2026Apr 22, 2026

    • fouroversix

      Public
      Code for the papers: “Four Over Six: More Accurate NVFP4 Quantization with Adaptive Block Scaling” and “Adaptive Block-Scaled Data Types”
      Python
      •
      MIT License
      •25•202•2•2•Updated Apr 21, 2026Apr 21, 2026

    • fastrl

      Public
      [ASPLOS'26] Taming the Long-Tail: Efficient Reasoning RL Training with Adaptive Drafter
      Python
      •
      Apache License 2.0
      •17•177•8•0•Updated Feb 27, 2026Feb 27, 2026

    • Block-Sparse-Attention

      Public
      A sparse attention kernel supporting mix sparse patterns
      C++
      •
      BSD 3-Clause "New" or "Revised" License
      •57•549•11•1•Updated Jan 18, 2026Jan 18, 2026

    • flash-moba

      Public
      C++
      •
      BSD 3-Clause "New" or "Revised" License
      •10•253•4•0•Updated Nov 19, 2025Nov 19, 2025

    • radial-attention

      Public
      [NeurIPS 2025] Radial Attention: O(nlogn) Sparse Attention with Energy Decay for Long Video Generation
      wanmochidiffusion-models
      wanmochidiffusion-modelssparse-attentionefficientmlhunyuan-video
      Python
      •
      Apache License 2.0
      •35•607•19•1•Updated Nov 11, 2025Nov 11, 2025

    • streaming-vlm

      Public
      StreamingVLM: Real-Time Understanding for Infinite Video Streams
      Python
      •
      MIT License
      •65•1.1k•27•0•Updated Oct 15, 2025Oct 15, 2025

    • efficientvit

      Public
      Efficient vision foundation models for high-resolution generation and perception.
      imagenetsegmentationhigh-resolution
      imagenetsegmentationhigh-resolutionvision-transformerefficientvitsegment-anythingdeep-compression-autoencoderefficient-diffusion-model
      Python
      •
      Apache License 2.0
      •256•3.4k•108•0•Updated Sep 5, 2025Sep 5, 2025

    • llm-awq

      Public
      [MLSys 2024 Best Paper Award] AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration
      Python
      •
      MIT License
      •323•3.6k•170•31•Updated Jul 17, 2025Jul 17, 2025

    • Quest

      Public
      [ICML 2024] Quest: Query-Aware Sparsity for Efficient Long-Context LLM Inference
      Cuda
      •
      MIT License
      •49•405•5•1•Updated Jul 10, 2025Jul 10, 2025

    • x-attention

      Public
      [ICML 2025] XAttention: Block Sparse Attention with Antidiagonal Scoring
      Python
      •24•283•10•0•Updated Jul 6, 2025Jul 6, 2025

    • vila-u

      Public
      [ICLR 2025] VILA-U: a Unified Foundation Model Integrating Visual Understanding and Generation
      Python
      •
      MIT License
      •19•428•19•0•Updated Apr 25, 2025Apr 25, 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 Spar…
      C++
      •
      Apache License 2.0
      •69•856•49•4•Updated Mar 6, 2025Mar 6, 2025

    • torchsparse

      Public
      [MICRO'23, MLSys'22] TorchSparse: Efficient Training and Inference Framework for Sparse Convolution on GPUs.
      accelerationpytorch
      accelerationpytorch
      Cuda
      •
      MIT License
      •193•1.5k•52•5•Updated Feb 24, 2025Feb 24, 2025

    • VisCompare

      Public
      A WebUI for Side-by-Side Comparison of Media (Images/Videos) Across Multiple Folders
      pythonuivideo
      pythonuivideoimageswebuidiffusion-modelsstreamlitgenerative-ai
      Python
      •
      Apache License 2.0
      •2•26•0•0•Updated Feb 21, 2025Feb 21, 2025

    • duo-attention

      Public
      [ICLR 2025] DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads
      Python
      •
      MIT License
      •42•539•14•1•Updated Feb 10, 2025Feb 10, 2025

    • patch_conv

      Public
      Patch convolution to avoid large GPU memory usage of Conv2D
      Python
      •
      MIT License
      •6•95•2•1•Updated Jan 23, 2025Jan 23, 2025

    • fastcomposer

      Public
      [IJCV] FastComposer: Tuning-Free Multi-Subject Image Generation with Localized Attention
      Python
      •
      MIT License
      •43•714•16•0•Updated Jan 10, 2025Jan 10, 2025

    • sparserefine

      Public
      [ECCV 2024] SparseRefine: Sparse Refinement for Efficient High-Resolution Semantic Segmentation
      Python
      •
      MIT License
      •1•16•1•0•Updated Jan 10, 2025Jan 10, 2025

    • distrifuser

      Public
      [CVPR 2024 Highlight] DistriFusion: Distributed Parallel Inference for High-Resolution Diffusion Models
      accelerationparallelismgenerative-model
      accelerationparallelismgenerative-modeldiffusion-modelsgenerative-ai
      Python
      •
      MIT License
      •34•728•11•1•Updated Dec 2, 2024Dec 2, 2024
    ProTip! When viewing an organization's repositories, you can use the props. filter to filter by custom property.