Local LLM on a pipeline of sparks

Companion app to control coding-agent harnesses from phone or desktop — OpenCode, PI, Oh My Pi (OMP) and Claude Code supported

Give your local LLM the beans!

Local-first, privacy-preserving multi-agent AI system for families. Runs on Raspberry Pi 5 + Jetson Orin NX. No cloud. No subscriptions. Your data stays home.

Structured OpenClaw update runbook for Codex/Claude operators: service checks, plugin drift, config regressions, task cleanup, and post-upgrade debugging.

Kernel-enforced authority and spend platform for AI agents

Reproducible token-throughput benchmark for OpenAI-compatible LLM servers, tuned for NVIDIA Spark and GB10 inference.

🦞 Low-latency, limited-context AI harness for private on-device homes.

Linux 0.11 rewritten in idiomatic Rust: kernel, std-style user library, and 60+ coreutils — boots on i386 in QEMU.

Foundation model for tiny devices; 14mb, 26m params, 1-6k toks/sec on mobiles, wearables smart home and robots.

Introducing Observa — a lightweight observability toolkit for collecting, transforming, and exporting traces & metrics.

An OpenCode plugin that lets AI agents control Windows desktops directly using the \windows-mcp\ MCP server.

Production-grade engineering skills for AI coding agents.

RegicideOS Arch, a BTRFS based OS with Cosmic Desktop and AI out of the box.

Unofficial Bitwarden compatible server written in Rust, formerly known as bitwarden_rs

Welcome syllabus and certification plan for the Vibe Coding Academy starter pack

My reasearch of losslessly compressing LLM weights.

CRS-自建Claude Code镜像，一站式开源中转服务，让 Claude、OpenAI、Gemini、Droid 订阅统一接入，支持拼车共享，更高效分摊成本，原生工具无缝使用。

Docker configuration for running VLLM on dual DGX Sparks

Uncensored local AI studio for Windows, Linux, and macOS. Zero-setup GUI for Image Generation, GGUF LLMs, Text to Speech & Speech to Text

Typed Cortex inspired by NormCode. - Governed grammar intelligence: intelligence whose primary act is to extend a formal grammar under governance constraints, preserving the provenance of every ext…

Directory of agentic skills, integrations, and tools for NVIDIA enterprise AI platforms.

Simulation-only defensive AI agent sandbox for studying persistent memory, auditable decision-making, proportional escalation, and human-supervised autonomous recommendations.

A KV cache CLI written in Rust

DIZI-AI is an open-source local AI orchestration platform for developers who want full control over their AI stack. Create intelligent multi-agent workflows, automate tasks, generate code and image…

Graph Application based on OpenCyc Ontology

Memory capacity planner for running LLMs on the NVIDIA DGX Spark: unified-memory budgeting, bandwidth-aware decode-speed estimates, and a quantization advisor.