Run larger LLMs with longer contexts on Apple Silicon by using differentiated precision for KV cache quantization. KVSplit enables 8-bit keys & 4-bit values, reducing memory by 59% with <1% quality loss. Includes benchmarking, visualization, and one-command setup. Optimized for M1/M2/M3 Macs with Metal support.

VGGT 3D Vision Agent optimized for Apple Silicon with Metal Performance Shaders

A Home Assistant Custom Component for Maestro MCZ (cloud)

Blender Addon to import and export m3 files

A port to Apple Silicon (see branch silicon)

🎨 Native AI image generation for Apple Silicon with Qwen-Image. Lightning LoRA acceleration for fast 4–8 step runs. Zero Docker, just works.

Material shapes in pure python.

A python training pipeline for Apple MLX to produce a fine tuned model with Apple Neural Engine Support

🎨 Generate AI images quickly on macOS with native CLI or containerized options for powerful, efficient results tailored for Apple Silicon.

An external script and template for getting Whatsminer M3 metrics into Zabbix

🍎 Accelerate 3D reconstructions from images using VGGT, optimized for Apple Silicon with Metal Performance Shaders for fast and efficient processing.

Anviz M3 RFID CVE-2019-11523 PoC

Libraries and utilities for interfacing with M3 and ICE

Download:

Visualizing m1 versus m3 traffic data with CesiumJS

🦑 KRAKEN - M3 Optimized SEFAZ Fiscal Analysis & Automation

LLM fine-tuning pipeline for Apple Silicon with MLX, web scraping, and LoRA training