Llm


Build Log

Two 128 GB AI Boxes

5-part series· updated

An NVIDIA DGX Spark and a Minisforum MS-S1 MAX (AMD Strix Halo) in one homelab: the hardware, what I run on them, a dated set of benchmarks on the same model, and how they end up working together, including the model tests and design that decide the roles.
Read the build log
Llama Rider
self-hosted

Llama Rider

· 1 min read

Autonomous AI agent runtime inspired by Open Claw, rebuilt from scratch to solve the problem of 100% Ollama compatibility—no OpenAI API shims, no …

Read more
Constitutional
self-hosted

Constitutional

· 1 min read

Tracks executive orders, federal rules, and congressional bills from the Federal Register and Congress.gov APIs, then runs each through an AI-powered …

Read more
Newsfeed
self-hosted

Newsfeed

· 1 min read

A personal news aggregator inspired by Kevin Rose’s Nylon project. Ingests articles from 800+ RSS feeds via FreshRSS, then uses AI embeddings to …

Read more