Nano vLLM

A high-throughput and memory-efficient inference and serving engine for LLMs

SQZ - Low complexity, scalable lossless and lossy image compression library

Developer Control Plane API server and CLI.

Run frontier MoE models on hardware you already own — pure C, zero deps, experts streamed from disk. Tiny engine, immense model. 🐦

Gas Town - multi-agent workspace manager

Autonomous Pokemon Red player that reads game memory, makes strategic decisions, and plays headlessly inside a stereOS VM, now with Kafka

Autonomous Pokemon Red player that reads game memory, makes strategic decisions, and plays headlessly inside a stereOS VM.

A powerful GPU profiling platform for Android Vulkan applications

Git - the stupid content tracker - This is a publish-only "broken out" repository and all pull requests are ignored. Please follow Documentation/SubmittingPatches procedure for any of your improvem…

TabFM (Tabular Foundation Model) is a pretrained tabular foundation model developed by Google Research for tabular data regression and classification.

A relative time formatting library, with no code.

LINGUISTIC ENGINE FOR AI

Lore is a next-generation, open source version control system

A C++ interpreter for the OPA policy language Rego

AI Agent Governance Toolkit — Policy enforcement, zero-trust identity, execution sandboxing, and reliability engineering for autonomous AI agents. Covers 10/10 OWASP Agentic Top 10.

Requirement-driven evaluation harness for AI agents and LLM applications. Generate behavior-specific test cases, run them against any target (hosted models, callable wrappers, OTel-traced agents), …

Universal Profiling as a desktop app

Python extension for generating profiling data and stack samples for Windows Performance Analyzer

An extensible host-based agent for Windows.

Automate the setup and configuration of your Windows development environment.

Reflect helps engineering teams see whether AI coding agents are actually improving delivery or just burning tokens, failing tools, and creating hidden debt

Lightweight, native Windows implementation of Linux commands | 1MB only | AI-friendly

User interface for recording and managing ETW traces

tiktoken is a fast BPE tokeniser for use with OpenAI's models.

The "We're doing our best" Platform-as-a-Service

Drop-in Acceleration of SQL/PromQL queries

An open-source project for Windows developers to learn how to add AI with local models and APIs to Windows apps.