omlx

MLX Studio - Home of JANG_Q - Image Gen/Edit + Chat/Code All in one - + OpenClaw (Anthropic API)

vMLX - JANGTQ Uber Compressed MLX Models - L2 Disk Cache (survives restart) + L1 Paged (super fast ttft) + Hybrid SSM Scheduler + Cont Batching + etc!

JANG — GGUF for MLX. YOU MUST USE JANG_Q RUNTIME. Adaptive Mixed-Precision Quantization + Runtime for Apple Silicon

Translate books and documents with LLMs, featuring glossary extraction, review, and polishing pipelines.

Community-driven benchmark suite for MLX inference engines on Apple Silicon

ZBoy - A sovereign, local coding agent on macOS.

Point it at a video, image, or PDF — get structured JSON. uvx vidlizer[mcp]. Runs local (Ollama/gemma4, LM Studio, oMLX) or cloud (OpenRouter). CLI + MCP server for Claude Code, Cursor, and Claude Desktop.

Real DuckDB Quack infrastructure for multi-agent Werewolf: containerized player nodes, Quack gateway federation, browser runner, and local/hosted LLM evals.

Memory-safe, LAN-accessible OpenAI-compatible server for Gemma 4 12B (4-bit MLX) on a 16 GB M2 Pro Mac mini

oMLX Prometheus exporter and Grafana dashboard

Frontier orchestration runtime for delegating work across harnesses, models, and providers.

Local AI on a Mac — oMLX inference, ds4, a Shorts maker, and launchd recipes for Apple Silicon

Intelligent task delegation for AI agents: Offload the "easy" work to your local machine and save premium tokens for the hard stuff.

Large Language Model governance and orchestration using ralph loops

Temporary Inkling Small compatibility patch for oMLX on Apple Silicon

Local AI on your terms — privacy-first native macOS, iOS, and iPadOS chat for Ollama, any OpenAI-compatible server (oMLX, Swama), and Apple Foundation Models. Nothing leaves your device. 🍎

bee coding agent harness

Local-first agentic terminal for private inference and extensible developer workflows

A MCP server to connect Claude Desktop to LMs on oMLX

Vendor-neutral benchmark for local LLM inference engines (omlx, llama.cpp): runs a fixed prompt set on your own hardware and reports real, measured tokens per second, not numbers from someone else's machine. CLI plus Python library, npm and PyPI, JSON output for CI and agents.