metal — independent software & tools
-
plank
Interactive terminal coding agent in Rust — runs the DeepSeek V4 Flash model locally on macOS (Metal), with a Ratatui TUI, a full tool suite, MCP, and persistent sessions.
-
damson
The terminal built only for macOS — GPU-accelerated (Swift + Metal + CoreText), Safari-smooth scrolling, flawless Korean input. Engine library + standalone Damson.app.
-
node-llama-cpp
Run AI models locally on your machine with node.js bindings for llama.cpp. Enforce a JSON schema on the model output on the generation level
-
Tournament lance
A medieval Lance made in about an hour, feel free to use it in your tornaments knights! 1K textures, midpoly model
-
Sci-fi Lab prop
a sci-fi lab prop for Colonization laboratory on another planet
-
stt-md
Lightweight macOS menubar app: mic → whisper → Ollama → Obsidian vault
-
squish
⚡️ The fastest way to run local LLMs on Apple Silicon — sub-second model loads, beats Ollama on throughput, tail latency, and full-response time. OpenAI/Ollama-compatible. No cloud, no API keys.
-
mold
CLI-native local AI image and video generation for people, scripts, and agents — CUDA on Linux, Metal on macOS, plus desktop, web, TUI, iPhone, REST/SSE, and MCP.