Search awesome repositories

《深入理解 AI Infra：量化分析与系统设计》（李博杰 著）开源书稿：从硬件约束和模型架构出发，量化推导 LLM 推理与训练系统设计。含全书正文、PDF、配套计算工具与实验

Achieve state of the art inference performance with modern accelerators on Kubernetes

eBPF Observability - Distributed Tracing and Profiling

🛠 All-in-one web-based IDE specialized for machine learning and data science.

Single C file, Realtime CPU/GPU Profiler with Remote Web Viewer

The Cross Platform Game Engine

High-performance inference framework for large language models, focusing on efficiency, flexibility, and availability.

2-2000x faster ML algos, 50% less memory usage, works on all hardware - new and old.

OpenLake is a high performance storage engine for efficient LLM inference and GPU Training

Vendor-agnostic orchestration for training, inference and agentic workloads across NVIDIA, AMD, TPU, and Tenstorrent on clouds, Kubernetes, and bare metal.

cuGraph - RAPIDS Graph Analytics Library

ThunderSVM: A Fast SVM Library on GPUs and CPUs

NVTabular is a feature engineering and preprocessing library for tabular data designed to quickly and easily manipulate terabyte scale datasets used to train deep learning based recommender systems.

Technical articles about console architecture

ThunderGBM: Fast GBDTs and Random Forests on GPUs

This library apply video filter on generate an Mp4 and on ExoPlayer video and Video Recording with Camera2.

CEED Library: Code for Efficient Extensible Discretizations

Open source math and physics engine that executes MATLAB-syntax code. Features a modern compiler, automatic GPU acceleration for large calculations, and support for math on geometry / FEA calculations.

Kubernetes operator for self-hosted LLM inference across a heterogeneous GPU fleet: NVIDIA CUDA, AMD Vulkan, and Apple Silicon Metal. Runtimes: llama.cpp, vLLM, TGI, mlx-server. Multi-GPU sharding, model caching, OpenAI-compatible endpoints. Apache-2.0, run across homelab and on-prem fleets, actively developed.

Plug-and-play homelab dashboard in one container — GPU, local-AI VRAM, Docker, systemd, host health. Built-in read-only MCP server so AI agents can explore it too.

AI agent builder with a VM at its heart. Every agent gets its own microVM sandbox to install software, run code and use the GPU in. Python, Node.js and Rust.

eBPF agent and MCP server for GPU causal observability

A dynamic binary instrumentation tool for tracing and analyzing GPU kernel instructions.

One-click deployment for OpenClaw + KasmVNC. Browser-based remote desktop with full container lifecycle management.

LLM-native terminal emulator for macOS — AI components, multi-provider profiles, voice input, Metal GPU rendering

Free, read-only GPU waste scanner for Kubernetes inference clusters, by Paralleliq. Finds idle GPUs, oversized GPU tiers and wasted capacity in vLLM deployments.

Holographic Reservoir Computing Framework (Simulation + Optional ASIC Hardware)

Abstract We present NeuroCHIMERA (Neuromorphic Cognitive Hybrid Intelligence for Memory-Embedded Reasoning Architecture), a novel GPU-native neuromorphic computing framework resulting from the integration of two complementary theoretical and computational advances:Veselov's Hierarchical Number System (HNS) with consciousness emergence parameters

GPU texture reservoir computing — fragment-shader cellular automata with linear readouts for image classification and PDE solving. Honest benchmarks, CPU/OpenGL/CUDA backends.

dockup - interactive TUI to install, harden and maintain Docker Engine, Compose v2, NVIDIA Container Toolkit and Apple container. One command, checksum-verified.