solatticus
GitHub
37 repos
8 followers
Explained projects
oMLX: The SSD-Backed Memory Tier for Apple Silicon
How a native macOS inference server uses tiered KV caching and continuous batching to eliminate the "Cold Start" problem for local agents.
8 min read · Mar 25, 2026
Yet to be explained
daisugi
Ternary QAT forge — bake any HuggingFace LLM down to ~2 bpw on your own training data.
Python
0 stars
Explain
utterance
Pipe text into a 3D space and walk through it. Markdown, ANSI color, images, keyboard navigation. Falls back to cat.
C
0 stars
Explain
