PKU-DAIR / PipeLink

A linker for partial GPU kernels.

Rust-GPU / rust-cuda

Ecosystem of libraries and tools for writing and executing fast GPU code fully in Rust.

unikraft / unikraft

A next-generation cloud native kernel designed to unlock best-in-class performance, security primitives and efficiency savings.

RenderKit / embree

Embree ray tracing kernels repository.

serge-sans-paille / pythran

Ahead of Time compiler for numeric kernels

CHIZI-0618 / box4magisk

Use sing-box, clash, v2ray, xray tunnel proxy on Android devices.

ashvardanian / NumKong

SIMD-accelerated distances, dot products, matrix ops, geospatial & geometric kernels for 16 numeric types — from 6-bit floats to 64-bit complex — across x86, Arm, RISC-V, and WASM, with bindings for Python, Rust, C, C++, Swift, JS, and Go 📐

m4rs-mt / ILGPU

ILGPU JIT Compiler for high-performance .Net GPU programs

almond-sh / almond

A Scala kernel for Jupyter

nanovms / ops

ops - build and run nanos unikernels

hermit-os / kernel

A Rust-based, lightweight unikernel.

twoone-3 / AdGuardHomeForRoot

A module to easily execute AdGuardHome on Android

Numbersf / Action-Build

Build SukiSU-Ultra/KernelSU/KernelSU-Next/BakaSU Kernels For All OnePlus QCOM & MTK Devices (GKI2.0), Supporting SUSFS, KPM, ZRAM and more feats.

lfortran / lfortran

Official main repository for LFortran

chelsea0x3b / cudarc

Safe rust wrapper around CUDA toolkit

flagos-ai / FlagGems

FlagGems is an operator library for large language models implemented in the Triton Language.

CrossGL / crosstl

Translate GPU shaders and compute kernels between GLSL, HLSL, Metal, SPIR-V and Slang, from CUDA and HIP, and to WGSL, through the CrossGL IR

emacs-jupyter / jupyter

An interface to communicate with Jupyter kernels.

Solo5 / solo5

A sandboxed execution environment for unikernels

jupyter-xeus / xeus

Implementation of the Jupyter kernel protocol in C++

NVIDIA / nvbench

CUDA Kernel Benchmarking Library

kernelslacker / trinity

Linux system call fuzzer

arighi / virtme-ng

Quickly build and run kernels inside a virtualized snapshot of your live system

WildKernels / OnePlus_KernelSU_SUSFS

OnePlus Kernels with KernelSU & SUSFS

okhsunrog / vpnhide

Hide active VPN from selected Android apps (kernel module + LSPosed + Zygisk)

NVIDIA / nvshmem

NVIDIA NVSHMEM is a parallel programming interface for NVIDIA GPUs based on OpenSHMEM. NVSHMEM can significantly reduce multi-process communication and coordination overheads by allowing programmers to perform one-sided communication from within CUDA kernels and on CUDA streams.

kernelsauce / turbo

Turbo is a framework built for LuaJIT 2 to simplify the task of building fast and scalable network applications. It uses a event-driven, non-blocking, no thread design to deliver excellent performance and minimal footprint to high-load applications while also providing excellent support for embedded uses.

zengxiao-he / tessera

From teacher to tiles — a from-scratch LLM distillation & serving engine: custom Triton/CUDA kernels, FSDP distillation, paged-KV continuous batching, speculative decoding, a Rust gateway, a JAX oracle, and interpretability tooling.

MANCrimSon / YouTube-ReVanced-Extended

Auto-built YouTube & YT Music (RVX/Morphe + VoT) with in-app updates. NonRoot APKs (Obtainium) and Root modules (Magisk/KernelSU/APatch).

kacper-daftcode / vLLM-Moet

A vLLM patch + hand‑written SM120 SASS kernels: 2‑bit MoE experts + an FP4 "delta" cache that recovers precision — matching the official (NV)FP4 checkpoint's quality on consumer Blackwell cards

maxsteeel / nomount

NoMount is a VFS (Virtual File System) path redirection framework for Android kernels.

dnv2003 / v100-skinny

Hand-written NVFP4 W4A16 CUDA kernels for Volta

Inferact / tpu-megakernels

A collection of megakernels for TPU

Oruk-AI / orukeet

Orukeet: multilingual ASR with fitted, frozen Gabor kernels and native inference

matsuzaka-yuki / susfs-lkm-module

KernelSU-compatible ZIP wrapper for pinned inforcqb SUSFS LKM; loader only, no rule management

script-or-script / FlutterTap

Zygisk module that intercepts Flutter app traffic — redirects to a proxy and bypasses BoringSSL TLS pinning. Persistent, no Frida session. Installable via MMRL.

yanght27 / GPU-Perf-Playground

GPU 性能与 AI Infra 学习项目：CUDA/Triton 算子、NCU/NSYS、vLLM/SGLang/TRT-LLM/ms-swift、PyTorch/DeepSpeed/ms-swift 训练、并行架构

mattn / tensai

A tiny neural-network framework in pure Go with AVX2 SIMD kernels (GOEXPERIMENT=simd)

IaroslavElistratov / gpu-kernel-engineering

B200 Attention from Scratch: understand one of the hardest GPU kernels on the latest hardware

julianmb / halofpx

High-performance Lemonade alternative for AMD Strix Halo & Radeon — latest MoE models (Ling-3.0-Flash, Qwen 3.8 Flash Next, Ornith 1.5), bleeding-edge RDNA 3.5 Wave64/ROCmFP4 kernels, and silicon-tuned profiles.

lyuyiqi / open-jev-fast

Faster inference backend for Open-Jev-27B: fused CUDA kernels, prefix tree, CUDA Graphs (B300, bf16)

poisonxa16 / pxa

The set-and-forget LLM engine for Pascal and Volta: PXQ codec + kernels, auto-tuned per card. Ready-to-run PXQ models in MODELS.md; benchmarks vs llama.cpp in the README.

lanyi233 / Sub-Store-Module

Sub-Store for Android (Magisk / KernelSU / APatch 模块)

NanoTurtle1145 / root-my-s24

Using CVE-2026-43499 to root your Galaxy S24 Series(SM-S92X0 ,(China / Hong Kong SAR / Taiwan))

hsfzxjy / npunlock

Write and run custom C kernels for Intel Core Ultra NPUs

Foxlape / GhostMode

Turn your Android device 'unavailable' for incoming calls while keeping LTE/5G mobile data fully working. Supports Shizuku & Root.

joshuaswarren / metal2vk

Compile Metal Shading Language compute kernels to Vulkan SPIR-V (clang C++ for OpenCL + a metal_stdlib shim + clspv)

antonellof / frink

Pure-Rust GGUF inference engine with quantized CPU, Metal & CUDA kernels, MoE support, OpenAI-compatible server. Benchmarked head-to-head against llama.cpp.

sky22333 / netboot-android

使Android手机能够充当USB 启动盘或PXE网络启动，Make an Android phone provide PXE network booting or a USB boot drive.

VisionR1 / TargetedHide

Ηide files, folders, and packages from specific target apps.

ShivamXD6 / ROM-Shifter

ROM Shifter is a an Android app built to make flashing, backing up, and migrating between custom ROMs as painless as possible. As well provides tools for some common things we do after switching to another ROM.

monovibe / s26u-m3q-temp-root

Temporary KernelSU root and LSPosed recovery app for Galaxy S26 Ultra SM-S948N AZG3

arpanpathak / gpu-parallel-book

📖 GPU parallel programming from the ground up : CUDA C++ and Rust 🦀 kernels paired with clickable SVG 2D/3D architecture diagrams (warps, SMs, memory tiers). Includes reduction, scan, matmul, and a full image‑processing capstone

ayyy7128 / ghostlock-xringo1

GhostLock XRING O1 root loader for validated 6.6.77 and 6.6.118 kernels

tsirysndr / bsdkrun

Instant MicroVMs/Unikernels for macOS and Linux

sudoingX / dgx-spark-deepseek-v41

DeepSeek-V4.1-Flash on two NVIDIA DGX Sparks: EXL3 weights from the official checkpoint, upstream vLLM with our patches and decode kernels

RightNow-AI / local-kimi

Optimized local serving engine for Kimi-Linear-48B: INT4 quantizer, fused decode kernels for a measured 3.18x, and an OpenAI-compatible server. Ships with k3, a bridge that detects the client per request so Claude Code, Codex, Cline, Aider and opencode all work unchanged.

nader295 / MaxManager

Advanced performance & power control for rooted Android — monitor and tune CPU, GPU, RAM, ZRAM, thermals, and more with Max AI and Max Atlas.

krishna3163 / best-root-apps-for-android

Curated directory of Android root apps, Magisk, KernelSU, LSPosed modules, system tools, privacy utilities, and no-root Shizuku apps.

sauravsingla / MemVanta

Low-memory C++20 LLM inference runtime for quantized GGUF models on CPU — mmap-backed weights, paged KV cache, Q4/Q8 kernels, and reproducible llama.cpp benchmarks.