Agent-first reverse engineering for macOS, Windows and Linux: any ACP agent reads, renames and patches binaries, Android apps and GPU kernels next to you.
A Kernel based root solution for Android
An advanced Kernel based root solution for Android
Automatic formal verification for warp-specialized GPU kernels.
Privacy focused NixOS pentesting environment with a customizable desktop and security tools.
ไธๅ 15T๏ผPLZ110๏ผไธดๆถ root + KernelSU per-boot ๆฅ็ฎก๏ผCVE-2026-64560 ็งปๆค๏ผไธๅๅๅบใ้ๅฏๅณๅคฑๆใๅ ๆ ธ 6.12.69-android16-6-g242e1a07f878ใ
FlashMLA: Efficient Multi-head Latent Attention Kernels
An Open-source book with Modern CUDA learning Notes for Beginners - BF16/FP8/FP4, HGEMM, CuTe, Flash-Attention, etc.
Standalone implementation of Zygisk
Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels
LMDeploy is a toolkit for compressing, deploying, and serving LLMs.
Open source inference engine that tunes its kernels on your hardware to find the fastest way to run each model. 2x faster than Ollama. Works on Apple Silicon, NVIDIA, AMD, or just a CPU.
Efficient Triton Kernels for LLM Training
Quantization, kernels, runtime and inference engine for mobiles, wearables, smart home and robots.
Discover best root apps, Magisk, KernelSu & LSPosed(xposed) modules & rooting guides
Ecosystem of libraries and tools for writing and executing fast GPU code fully in Rust.
Transparent implementation of Zygisk.
A next-generation cloud native kernel designed to unlock best-in-class performance, security primitives and efficiency savings.
A KSU WebUI to configure Tricky Store target.txt
A kernel designed to run one and only one application in a virtualized environment
LLM speculative inference server for heterogeneous hardware & consumer GPUs
่ชๅปบๆผ้ณ่พๅ ฅๆณ่ฏๅบ๏ผ็พไธๅธธ็จ่ฏๆฑ้๏ผ้้ Fcitx5 (Linux / Android) ๅ Gboard (Android + Magisk or KernelSU) ใ
Embree ray tracing kernels repository.
A toolkit for managing Play Integrity & System Environment
An addon root hiding service for KernelSU
CUDA Core Compute Libraries
A lightweight, LXC-like container runtime for Android and Linux. Run full Linux distributions natively with zero performance penalty
A app for managing, sharing, and exploring Magisk modules across repositories
cuTile is a programming model for writing parallel kernels for NVIDIA GPUs
Ahead of Time compiler for numeric kernels
A kernel library written in tilelang
A KernelSU based root solution for Android
Use sing-box, clash, v2ray, xray tunnel proxy on Android devices.
SIMD-accelerated distances, dot products, matrix ops, geospatial & geometric kernels for 16 numeric types โ from 6-bit floats to 64-bit complex โ across x86, Arm, RISC-V, and WASM, with bindings for Python, Rust, C, C++, Swift, JS, and Go ๐
ILGPU JIT Compiler for high-performance .Net GPU programs
GKI2 5.10+ Kernels built with KernelSU & SUSFS
:metal: TT-NN operator library, and TT-Metalium low level kernel programming model.
A Scala kernel for Jupyter
linux-tkg custom kernels
ops - build and run nanos unikernels
A Rust-based, lightweight unikernel.
A module to easily execute AdGuardHome on Android
Systemless hosts for APatch / KernelSU / Magisk
Globally mounted modules via OverlayFS.
Kernel Design Agents (KDA) is a agent-centric workflow to write high-performance CUDA Kernels.
Build SukiSU-Ultra/KernelSU/KernelSU-Next/BakaSU Kernels For All OnePlus QCOM & MTK Devices (GKI2.0), Supporting SUSFS, KPM, ZRAM and more feats.
Based on the sing-box core, this Android proxy module supports one-click start/stop of transparent proxy and is designed for Android devices.
Official main repository for LFortran
๐ A ranked list of awesome Jupyter Notebook, Hub and Lab projects (extensions, kernels, tools). Updated weekly.
Safe rust wrapper around CUDA toolkit
A Quirky Assortment of CuTe Kernels
FlagGems is an operator library for large language models implemented in the Triton Language.
Translate GPU shaders and compute kernels between Metal, HLSL and GLSL, and to SPIR-V and WGSL, through the CrossGL IR
An interface to communicate with Jupyter kernels.
A sandboxed execution environment for unikernels
cuDNN Frontend is NVIDIA's modern, open-source entry point to the cuDNN library and a growing collection of high-performance open-source kernels.
A Python-embedded DSL that makes it easy to write fast, scalable ML kernels with minimal boilerplate.
Implementation of the Jupyter kernel protocol in C++
CUDA Kernel Benchmarking Library
Linux system call fuzzer