ModemGuides BlogGuides for people who would rather own their hardware than rent it.
AI Infrastructure · Topic
Every AI Infrastructure article tagged Local AI, newest first.
18 articles
Comparison
Apple's 2026 Mac mini refresh puts the newest M6 chip in the cheaper model and the M5 Pro in the expensive one, which makes the choice less obvious than it looks. For local AI the answer comes down to two numbers: how...
Reality check
Qwen 4 27B was named on stage September 22, not shipped. The Qwen3.8-27B hardware baseline, what the new architecture could change, and what to wait for.
Reality check
V4.1-Flash is 552B and ships at 510GB with experts already in FP4, so quantization cannot rescue it. The tier that holds it, and what to run now.
Buying guide
Find your memory number and get the one verified pick for it. Six tiers from a 12GB card to a 512GB Mac Studio, every size sourced from the repository, every license checked, re-verified September 5, 2026.
Reality check
GLM-5.3-Flash lands at 320B with 18B active under MIT. The honest hardware math: GGUF sizes, the 128GB line, and day-one runtime status.
VRAM
The 0731 weights are out and the tier map is current: 103GB build on mainline llama.cpp, LM Studio gone local, and the M5 Mac Studio math.
Buying guide
Which GPU and how much VRAM for local AI in 2026 — honest build tiers repriced for the memory shortage, from used RTX 3090 to Apple Silicon.
Buying guide
Apple's Mac Studio M5 Ultra is official: 512GB max, not 768GB. Which local AI models fit, the cluster math, and whether to buy, wait, or skip.
Memory prices
Nvidia's Rubin racks are shipping at scale, each pooling 20.7TB of HBM4. The honest buy-now-or-wait call for GPUs, RAM, and SSDs in late 2026.
Reality check
Open models now sit in the frontier top five, but half the viral scoreboard is unshipped. The verified ledger, the price math, and what runs at home.
Memory prices
The viral RAND chart, explained without the geopolitics: why AI's power and chip limits are already raising your costs, and the local-first hedge.
How-to
A self-verifying AI agent loop checks its own work against live sources. Here is the real technique behind the viral Kimi swarm, minus the hype.
Reality check
The honest tokens-per-second, the $1,499-vs-$2,200 SKU trap, the Linux memory unlock, and whether a Strix Halo box can replace your Claude Code bill.
Buying guide
RAM prices tripled, Apple pulled high-memory Macs, and relief isn't coming in 2026. Here's the shortage explained — and how to build local AI anyway.
Security
AI models now find zero-days in every major OS. Karpathy's 15-step guide plus home network hardening to protect your digital life in 2026.
Andrej Karpathy's LLM Wiki pattern lets an AI build and maintain a structured knowledge base from your documents. Here's how to set it up locally with Obsidian and Ollama — no cloud services, no data leaving your network.
Agents
AutoAgent proves AI agents can autonomously modify their own tools and behavior. What self-improving agents mean for your local AI stack and home network security.
How-to
Build a private, local-first AI knowledge base using Obsidian, Ollama, and AnythingLLM. Your research stays on your network with zero monthly fees.