Stars
An online request replication and TCP stream replay tool, ideal for real testing, performance testing, stability testing, stress testing, load testing, smoke testing, and more.
A collaborative list of awesome Zig libraries and resources.
A curated resource list for learning AI performance engineering, from GPU fundamentals to production inference.
The most atomic way to train and inference a GPT in pure, dependency-free C
100 days of LLM inference engineering — daily posts, experiments, and visualizations
Reference implementation and examples of the CuTe Layout representation and algebra.
simple triangle displayed using vulkan, xcb, and zig
Techniques and numbers for estimating system's performance from first-principles
A collection of resources for learning type theory and type theory adjacent fields.
Information about many aspects of high-performance computing. Wiki content moved to ~/docs.
mKernel: fast multi-node, multi-GPU fused kernels
This module defines a type system for distributed training code, based off of JAX's sharding in types, but adapted for the PyTorch ecosystem.
cuda-oxide is a Rust-to-CUDA compiler that lets you write (SIMT) GPU kernels in safe(ish), idiomatic Rust. It compiles standard Rust code directly to PTX — no DSLs, no foreign language bindings, ju…
DeepSeek 4 Flash and PRO local inference engine for Metal, CUDA and ROCm
Any model. Any hardware. Zero compromise. Built with @ziglang / @openxla / MLIR / @bazelbuild
Tilus is a tile-level kernel programming language with explicit control over shared memory and registers.
A tiny educational decompiler that helps people understand how decompilation works.
An experimental implementation of compiler-driven automatic sharding of models across a given device mesh.
LLM speculative inference server for heterogeneous hardware & consumer GPUs
Method for Long Context RLMs using verifiable Lambda Calculus
This is an online course where you can learn and master the skill of low-level performance analysis and tuning.
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 te…
A hardware-aware guide to data structures for system software engineers.
Trio – a friendly Python library for async concurrency and I/O