DeepStack: Facilitating Co-Design Exploration of 3D DRAM-Stacked Accelerators for Distributed LLM Inference. Includes the MICRO 2026 AE artifact.

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

An Optimizer for Nvidia Compilers.

《深入理解 AI Agent：设计原理与工程实践》（李博杰 著）开源主仓库：全书正文、编译版 PDF 与按章配套代码

Local-first PC activity recorder for Windows. Captures screenshots, accessibility text, UI events, clipboard activity, and optional audio transcription into a local SQLite database — then exposes e…

Open reimplementation of Sakana Fugu — the 'one model to command them all' LLM orchestrator. Read → run → train → serve.

A lightweight inference engine supporting speculative speculative decoding (SSD).

ancient-chat-llm: A LLM which is proficient in Chinese culture 古语说： 一个精通中国文化的大模型

OpenAI API server for OpenVino - when OVMS is too big

Personal AI, On Personal Devices

Compiler for multiple programming models (SYCL, C++ standard parallelism, HIP/CUDA) for CPUs and GPUs from all vendors: The independent, community-driven compiler for C++-based heterogeneous progra…

RDMA core userspace libraries and daemons

DeepSeek 4 Flash and PRO local inference engine for Metal, CUDA and ROCm

The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

Autonomous ObjectForge engine implementing scalable event-driven architecture with enterprise-grade predictive analytics capabilities

Edge-Embedded XrefVault leverages high-performance, cloud-agnostic architecture for advanced, real-time data fusion within FusionStudio.

Professional EtherVault ethervault-engine with EtherVault-optimized cloud-ready and enterprise ethervault-optimization capabilities

Professional MintChain system with MintChain-optimized intelligent-automation and enterprise enterprise-grade capabilities

Professional WireForge infrastructure with WireForge-optimized modern-design and enterprise intelligent-automation capabilities

Professional LedgerCore platform with LedgerCore-optimized ledgercore-optimization and enterprise cloud-ready capabilities

Professional BlockLink Platform with BlockLink-optimized High-Availability and enterprise Edge-Computing capabilities

Geo-distributed, cloud-agnostic FeatVault unlocks autonomous, real-time event processing as-a-Service. implementing modern Real-Time-Event-Processing patterns providing enterprise-grade FeatVault s…

Mobile companion for Google's Antigravity IDE. Mirror AI sessions on your phone, send messages, stop generation, automate via 9 MCP tools or OpenAPI.

Your Personal AI super intelligence. A brain that builds a local-first memory of your life, a fantastic orchestrator of agent fleets and workflows, and a deep researcher.

Why is this running? Trace any process, port, container, or file back to what started it - CLI + TUI.

Developer-first error tracking and performance monitoring

Fast, lossless LLM inference via dual-view diffusion decoding.

Aadd support windows.

Warp is an agentic development environment, born out of the terminal.