╔══════════════════════════════════════════════════════════╗
║ PERSONAL LARGE LANGUAGE LISP MACHINE · SNAPKITTYWEST ║
╚══════════════════════════════════════════════════════════╝
LLSM BUILDER READY.
Select your blocks in the left panel, then hit BOOT.
Your machine. Your model. Your memory. No cloud required.
MODE: LOCAL/OLLAMA
· MODEL: none selected
· KERNEL: RAW
· MEMORY: WORM
NAME: —
WORM CHAIN · SESSION LEDGER
— No entries yet. Boot your LLSM to start the chain. —
Boot your LLSM first to start talking. →
All Free Nano LLMs
— runs via Ollama on your hardware
How to run LLSM on your hardware
1
Install Ollama
Ollama runs open-source LLMs locally. Free. No account needed.
→ Download from ollama.ai
Available for Mac, Windows, Linux. Runs on CPU if no GPU available.
2
Enable CORS (required for browser access)
OLLAMA_ORIGINS="*" ollama serve
Or set permanently: add OLLAMA_ORIGINS=* to your shell profile.
3
Pull your first nano model
ollama pull llama3.2:3b
ollama pull deepseek-r1:1.5b
ollama pull nemotron-mini:4b
Browse the full catalog in the MODEL CATALOG tab. All models are free and open-source.
4
Boot this page and select LOCAL mode
The LLSM Builder connects to http://localhost:11434 automatically.
Select LOCAL mode → pick your model → hit BOOT.
5
No GPU? No problem.
Use CLOUD mode with a free Groq API key (groq.com) — fastest free inference available.
Or use DEMO mode to explore the builder without any setup.
Groq free tier: 6,000 requests/day · llama3-8b-8192 · 0ms cold start
★
Download APPLE GITDOS to run everything offline
Download the entire Apple GITDOS environment — LLSM Builder, twin dashboard,
WORM ledger, all tools — to run on your own machine or private server.
↓ DOWNLOAD APPLE GITDOS (ZIP)