gpu-verifier / sentryward
Automatic formal verification for warp-specialized GPU kernels.
gpu-verifier / sentryward
SerenityOS / serenity
The Serenity Operating System 🐞
deepseek-ai / FlashMLA
FlashMLA: Efficient Multi-head Latent Attention Kernels
cactus-compute / cactus
Quantization, kernels, runtime and inference engine for mobiles, wearables, smart home and robots.
ROCm / hip
HIP: C++ Heterogeneous-Compute Interface for Portability
abcz316 / SKRoot-linuxKernelRoot
新一代 SKRoot，完美隐藏Root功能，无视全网检测手段，实现SELinux零触碰、无挂载！ 通杀所有内核，免源码直接 Patch 原厂内核，完美保留官方内核所有特性。
Luce-Org / lucebox
LLM speculative inference server for heterogeneous hardware & consumer GPUs
RenderKit / embree
Embree ray tracing kernels repository.
NVIDIA / cccl
CUDA Core Compute Libraries
Idov31 / Nidhogg
Windows rootkit for Intel x64 with 25+ features, demonstrating rootkit techniques compatible with all Windows 10 and Windows 11 versions.
serge-sans-paille / pythran
Ahead of Time compiler for numeric kernels
managarm / managarm
Pragmatic microkernel-based OS with fully asynchronous I/O
lfortran / lfortran
Official main repository for LFortran
NVIDIA / cuda-tile
CUDA Tile IR is an MLIR-based intermediate representation and compiler infrastructure for CUDA kernel optimization, focusing on tile-based computation patterns and optimizations targeting NVIDIA tensor core units.
jupyter-xeus / xeus
Implementation of the Jupyter kernel protocol in C++
opengauss-mirror / openGauss-server
openGauss kernel
vxlang / vxlang-page
protector & obfuscator & code virtualizer
acidanthera / AirportBrcmFixup
An open source kernel extension providing a set of patches required for non-native Airport Broadcom Wi-Fi cards.
flashrt-project / FlashRT
FlashRT is a high-performance realtime inference engine for small-batch, latency-sensitive AI workloads. The flagship integration is production VLA control for Pi0, Pi0.5, GROOT N1.6, and Pi0-FAST. Also support llm e.g, qwen3.6-27B
Xilinx / XRT
Run Time for AIE and FPGA based platforms
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.
deepseek-ai / DeepGEMM-Ascend
DeepGEMM-Ascend: clean and efficient matrix multiplication kernel library for Huawei Ascend NPUs
KSwordDEV / KSword
[Windows Anti-Rootkit] KSword 5.1 is an source-available Windows toolkit for ARK, kernel debugging, and system forensics. KSword 5.1 是面向 Windows 的源码可见 ARK、内核调试与系统取证工具集。
ROCm / composable_kernel
[DEPRECATED] Moved to ROCm/rocm-libraries repo. NOTE: develop branch is maintained as a read-only mirror
deepseek-ai / DeepJIT
A lightweight library for xPU kernel JIT compilation
snuri00 / psp-web-recomp
PSP games in the browser, static recompilation to WebAssembly, an HLE kernel and a WebGL2 renderer
OnyxZygisk / OnyxZygisk
Zygote injection via ptrace - Zygisk API for APatch/KernelSU/Magisk, with FN (Functional Node) modules and a webroot WebUI
Almosst-DEV / Cobalt64
Cobalt64: native macOS GPU acceleration for unsupported AMD Radeon GPUs. First supported family: RDNA4 (RX 9070 XT / Navi 48). Kernel driver, RADV Darwin port, Metal-on-Vulkan bundle.
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.
cohere-ai / cohere-megakernel
—
XiaoMiku01 / TetherKit
Kext-free HoRNDIS alternative: Android USB tethering on macOS. A user-space RNDIS driver for Apple Silicon & Intel — no kernel extension, no SIP changes. GUI + CLI.
winterknife / PLATINUMPICK
Windows Kernel-Mode Shellcode Development Framework (WKMSDF)
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.
Rurixis / MSRKit
A concept of using a ROP chain paired with a WRMSR primitive to call kernel functions and map unsigned drivers through BYOVD (AmdTools64.sys)
vermellcc / vermell
Minimal, dependency-free C++ web framework using Linux epoll. (Revived & audited from our archived legacy codebase).
VisionR1 / TargetedHide
Ηide files, folders, and packages from specific target apps.
Shinn-Home / InvisibleHook
Invisible hook code(support win10 17134 ~ win11 26H1 ) InvisibleHook is an experimental Windows kernel/user-mode research project.
nbs32k / LocalStranger
PoC for WinNotify, demonstrated through a driver mapper, and local privilege escalation.
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.
Geramy / LSE
HRX, LOOM, Mac, and AMD running top speed kernels with a self optimizing engine.
cair-vinuni / vla.simd
A pure C++ inference engine for VLA policies on CPUs, with no GPU, CUDA, or ggml dependency. Built with its own tensors, operators, and SIMD kernels, the engine keeps kernels easy to inspect, tune, and replace.
aaalll12322 / ComfyUI-MiniMaxH3-SolAttn-V100
Sol-Attn (arXiv 2607.24027) sparse acceleration for MiniMax H3 on V100 — single-node ComfyUI plugin (embedded FP16Safe + keep-or-drop kernel)
TonyJoaca / DLSSG-Transfusion
Universal DLSS Multi-Frame Generation enabler (2x through 6x and Dynamic Mode) for NVIDIA RTX 40-Series (Ada Lovelace) GPUs with runtime Blackwell Kernel Transfusion.
Applet-LLC / OpenInputBridge
Clean-room, original-protocol-compatible reimplementation of the Interception keyboard/mouse kernel driver's IOCTL protocol for Windows 11.