Skip to content

@madsys-dev MadSys Research Group

Change the repository type filter

All

    Repositories list

    45 repositories

    • sglang

      Public
      SGLang is a fast serving framework for large language models and vision language models.
      Python
      •
      Apache License 2.0
      •7.3k•1•0•5•Updated Jul 16, 2026Jul 16, 2026

    • tokenspeed

      Public
      TokenSpeed is a speed-of-light LLM inference engine.
      Python
      •
      MIT License
      •192•1•0•0•Updated May 16, 2026May 16, 2026

    • smg

      Public
      Engine-agnostic LLM gateway in Rust. Full OpenAI & Anthropic API compatibility across SGLang, vLLM, TRT-LLM, OpenAI, Gemini & more. Industry-first gRPC pipeline…
      Rust
      •
      Apache License 2.0
      •122•0•0•0•Updated May 3, 2026May 3, 2026

    • DeepGEMM

      Public
      DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling
      Cuda
      •
      MIT License
      •1.1k•0•0•0•Updated Apr 24, 2026Apr 24, 2026

    • flashinfer

      Public
      FlashInfer: Kernel Library for LLM Serving
      Python
      •
      Apache License 2.0
      •1.2k•0•0•0•Updated Apr 22, 2026Apr 22, 2026

    • sglang-2

      Public
      SGLang is a high-performance serving framework for large language models and multimodal models.
      Python
      •
      Apache License 2.0
      •7.3k•1•0•0•Updated Apr 3, 2026Apr 3, 2026

    • xgrammar

      Public
      Fast, Flexible and Portable Structured Generation
      C++
      •
      Apache License 2.0
      •172•0•0•0•Updated Mar 27, 2026Mar 27, 2026

    • TorchSpec

      Public
      A PyTorch native library for training speculative decoding models
      Python
      •
      MIT License
      •49•0•0•0•Updated Mar 21, 2026Mar 21, 2026

    • overlaybd

      Public
      Overlaybd: a block based remote image format. The storage backend of containerd/accelerated-container-image.
      C++
      •
      Apache License 2.0
      •76•0•0•0•Updated Dec 24, 2025Dec 24, 2025

    • accelerated-container-image

      Public
      A production-ready remote container image format (overlaybd) and snapshotter based on block-device.
      Go
      •
      Apache License 2.0
      •89•0•0•0•Updated Dec 7, 2025Dec 7, 2025

    • Mooncake

      Public
      Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI.
      C++
      •
      Apache License 2.0
      •979•0•0•0•Updated Sep 29, 2025Sep 29, 2025

    • rbg

      Public
      A workload for deploying LLM inference services on Kubernetes
      Go
      •
      Apache License 2.0
      •69•0•0•0•Updated Sep 23, 2025Sep 23, 2025

    • checkpoint-engine

      Public
      Checkpoint-engine is a simple middleware to update model weights in LLM inference engines
      Python
      •
      MIT License
      •97•0•0•0•Updated Sep 22, 2025Sep 22, 2025

    • DSA-2LM

      Public
      The Artifact Evaluation Version of ATC'25 Paper #395 DSA-2LM: A CPU-Free Tiered Memory Architecture with Intel DSA
      C
      •0•9•0•0•Updated Jul 8, 2025Jul 8, 2025

    • scalio-osdi25-ae

      Public
      This is the repository for our OSDI '25 paper: Scalio: Scaling up DPU-based JBOF Key-value Store with NVMe-oF Target Offload.
      C
      •1•5•0•0•Updated May 23, 2025May 23, 2025

    • Semi-PD

      Public
      A prefill & decode disaggregated LLM serving framework with shared GPU memory and fine-grained compute isolation.
      Python
      •
      Apache License 2.0
      •16•0•0•0•Updated May 17, 2025May 17, 2025

    • async-ucx

      Public
      Asynchronous Rust bindings for UCX
      rustnetworkbindings
      rustnetworkbindingsucx
      Rust
      •
      MIT License
      •20•80•2•2•Updated Apr 29, 2025Apr 29, 2025

    • deepseekv2-profile

      Public
      Jupyter Notebook
      •19•156•3•2•Updated Mar 4, 2025Mar 4, 2025

    • sosp-paper19-ae

      Public
      The Artifact Evaluation Version of SOSP Paper #19
      C++
      •13•54•1•2•Updated Aug 19, 2024Aug 19, 2024

    • VertexSurge

      Public
      C++
      •0•5•0•0•Updated Jun 25, 2024Jun 25, 2024

    • smart

      Public
      Scaling Up Memory Disaggregated Applications with SMART
      C++
      •
      MIT License
      •7•35•1•0•Updated Apr 23, 2024Apr 23, 2024

    • vllm

      Public
      A high-throughput and memory-efficient inference and serving engine for LLMs
      Python
      •
      Apache License 2.0
      •20k•0•0•0•Updated Dec 23, 2023Dec 23, 2023

    • Falcon

      Public
      Falcon: Fast OLTP Engine for Persistent Cache and Non-Volatile Memory
      Rust
      •
      MIT License
      •0•11•1•0•Updated Nov 1, 2023Nov 1, 2023

    • NBTree

      Public
      NBTree for Falcon (fork from https://github.com/SJTU-DDST/NBTree)
      C++
      •0•0•0•0•Updated Aug 22, 2023Aug 22, 2023

    • dash

      Public
      Dash for Falcon (forked from https://github.com/baotonglu/dash)
      C
      •
      MIT License
      •1•0•0•0•Updated Jul 27, 2023Jul 27, 2023

    • ROART

      Public
      Range Optimized Adaptive Radix Tree
      C++
      •
      Other
      •13•25•2•0•Updated Feb 1, 2023Feb 1, 2023

    • MadEngine

      Public
      MadFS local storage engine
      Rust
      •3•6•0•0•Updated Nov 7, 2022Nov 7, 2022

    • async-spdk

      Public
      Asynchronous Rust bindings for SPDK.
      rustasyncbindings
      rustasyncbindingsspdk
      Rust
      •4•18•0•0•Updated Nov 1, 2022Nov 1, 2022

    • syscall_intercept

      Public
      The system call intercepting library
      C
      •
      Other
      •130•23•0•0•Updated Sep 18, 2022Sep 18, 2022

    • syscall-intercept-rs

      Public
      A Rust wrapper of pmem/syscall_intercept.
      Rust
      •
      MIT License
      •1•7•0•0•Updated Jun 23, 2022Jun 23, 2022
    ProTip! When viewing an organization's repositories, you can use the props. filter to filter by custom property.