Skip to content

@sgl-project sgl-project

Change the repository type filter

All

    Repositories list

    25 repositories

    • sglang

      Public
      SGLang is a high-performance serving framework for large language models and multimodal models.
      reinforcement-learningcudainference
      reinforcement-learningcudainferencetransformermoeattentionllamaglmminimaxwan
      Python
      •
      Apache License 2.0
      •5k•25k•596•1.9k•Updated Mar 23, 2026Mar 23, 2026

    • sglang-jax

      Public
      JAX backend for SGL
      Python
      •
      Apache License 2.0
      •76•252•90•33•Updated Mar 23, 2026Mar 23, 2026

    • sgl-project.github.io

      Public
      This is the documentation repository for SGLang. It is auto-generated from https://github.com/sgl-project/sglang
      HTML
      •31•115•11•1•Updated Mar 23, 2026Mar 23, 2026

    • ome

      Public
      Open Model Engine (OME) — Kubernetes operator for LLM serving, GPU scheduling, and model lifecycle management. Works with SGLang, vLLM, TensorRT-LLM, and Triton
      k8sllamaoracle-cloud
      k8sllamaoracle-cloudmodel-servingmodel-as-a-servicemulti-node-kubernetesllmvllmllm-inferenceqwen
      Go
      •
      Apache License 2.0
      •66•401•32•46•Updated Mar 23, 2026Mar 23, 2026

    • sgl-docs

      Public
      MDX
      •
      Apache License 2.0
      •16•4•0•0•Updated Mar 23, 2026Mar 23, 2026

    • rbg

      Public
      A workload for deploying LLM inference services on Kubernetes
      k8sllmsglang
      k8sllmsglangpd-disagg
      Go
      •
      Apache License 2.0
      •49•192•25•25•Updated Mar 23, 2026Mar 23, 2026

    • sgl-kernel-xpu

      Public
      SGLang kernel library for Intel XPU
      Python
      •
      MIT License
      •21•20•0•16•Updated Mar 23, 2026Mar 23, 2026

    • srt-slurm

      Public
      Benchmark SGLang on SLURM
      Python
      •37•0•0•0•Updated Mar 23, 2026Mar 23, 2026

    • whl

      Public
      SGLang Kernel Wheel Index
      cudacutlasssglang
      cudacutlasssglangflashinfer
      HTML
      •
      MIT License
      •9•18•0•1•Updated Mar 23, 2026Mar 23, 2026

    • sgl-cookbook

      Public
      Cookbook of SGLang - Recipe
      JavaScript
      •
      Apache License 2.0
      •48•99•6•7•Updated Mar 22, 2026Mar 22, 2026

    • sglang-omni

      Public
      SGLang Omni: High-Performance Multi-Stage Pipeline Framework for Omni Models
      Python
      •
      MIT License
      •26•96•43•25•Updated Mar 22, 2026Mar 22, 2026

    • SpecForge

      Public
      Train speculative decoding models effortlessly and port them smoothly to SGLang serving.
      trainingeaglepytorch
      trainingeaglepytorchllmfsdpsglangeagle3
      Python
      •
      MIT License
      •186•741•65•39•Updated Mar 21, 2026Mar 21, 2026

    • rbg-api

      Public
      1•0•0•1•Updated Mar 19, 2026Mar 19, 2026

    • genai-bench

      Public
      A Rust reimplementation of genai-bench for benchmarking LLM serving systems at high concurrency with accurate timing and industry-standard metrics
      Python
      •
      MIT License
      •50•284•9•9•Updated Mar 18, 2026Mar 18, 2026

    • sgl-kernel-npu

      Public
      SGLang kernel library for NPU
      C++
      •
      MIT License
      •96•108•15•43•Updated Mar 18, 2026Mar 18, 2026

    • DeepGEMM

      Public
      DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling
      Cuda
      •
      MIT License
      •840•22•0•1•Updated Mar 18, 2026Mar 18, 2026

    • sgl-flash-attn

      Public
      Fast and memory-efficient exact attention
      Python
      •
      BSD 3-Clause "New" or "Revised" License
      •2.5k•21•0•0•Updated Mar 13, 2026Mar 13, 2026

    • mini-sglang

      Public
      A compact implementation of SGLang, designed to demystify the complexities of modern LLM serving systems.
      Python
      •
      MIT License
      •517•3.8k•8•24•Updated Mar 13, 2026Mar 13, 2026

    • sgl-test-files

      Public
      The test files for SGLang.
      MIT License
      •3•1•0•1•Updated Feb 23, 2026Feb 23, 2026

    • FlashMLA

      Public
      FlashMLA: Efficient Multi-head Latent Attention Kernels
      C++
      •
      MIT License
      •1k•0•0•0•Updated Feb 16, 2026Feb 16, 2026

    • gb200-ci-runner-public

      Public
      0•0•0•0•Updated Jan 20, 2026Jan 20, 2026

    • ome-crd

      Public
      0•0•0•0•Updated Jan 15, 2026Jan 15, 2026

    • sgl-learning-materials

      Public
      Materials for learning SGLang
      MIT License
      •60•785•0•0•Updated Jan 5, 2026Jan 5, 2026

    • fast-hadamard-transform

      Public
      Fast Hadamard transform in CUDA, with a PyTorch interface
      C
      •
      BSD 3-Clause "New" or "Revised" License
      •56•1•0•0•Updated Oct 15, 2025Oct 15, 2025

    • sgl-whl

      Public
      SGLang wheels for multiple platforms
      MIT License
      •2•11•1•0•Updated Oct 13, 2025Oct 13, 2025