Havuz

Native Rapid-MLX provider for DeepSeek Harness (dsh) — dsh reads model facts from the server instead of your settings.yaml.

A practical lab for exploring Apple's Core AI framework, model assets, specialization, and on-device inference.

Big models. Small Macs. Zero excuses.

TurboQuant MLX: Flagship high-performance model quantization for Apple Silicon. Squeeze maximum inference speed and memory efficiency out of

Community benchmark database for running LLMs on Apple Silicon Macs

A desktop app for creating and managing isolated Mac environments.

Offline emergency assistant: a local LLM and a 200 GB offline library on one Mac, with cited answers and no internet.

Local, private push-to-talk dictation for macOS. Hold a key, speak, release — your words appear in any text field. Runs entirely on-device v

Macaw — on-device AI agent for macOS. 2.7B LFM2.5 fine-tune, MLX 4-bit, 97 macOS tools, no cloud.

Fast, private push-to-talk dictation for Apple Silicon Macs — local Parakeet TDT v3, no cloud or telemetry.

Native AI assistant for macOS, built with SwiftUI and llama.cpp. Run local LLMs offline, work with PDFs, and create custom AI assistants. Op

Fast, private, on-device dictation for Apple silicon Macs

AI watermark removal built for Apple Silicon. SynthID-class pixel regeneration, visible vendor labels and provenance metadata, with every pi

Free, open source dictation for Mac. A Wispr Flow alternative that runs offline and grows with TypeScript extensions.

The first native Swift inference engine for Gemma 4 on Apple Silicon / iOS

Local AI workstation: chat by voice or text, transcribe, generate images and video (with sound on Apple Silicon), benchmark and serve open-w

Local LLM chat app for Apple Silicon Macs.

🧠 One LLM, split across a Mac (Apple MPS) + a Windows PC (NVIDIA CUDA) — heterogeneous pipeline inference over a framework-neutral wire, bit

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

Open-source macOS AI computer use and native GUI automation on Apple silicon. Jev + OmniParser CoreML + Apple Vision OCR. Bring your own Ope

On-device live transcription and translation CLI for macOS 26+ on Apple Silicon

Transformer inference via unconventional compute: a single algebraic primitive, a tag-system scheduler, and Apple's Neural Engine through co

Native macOS sidebar terminal that reveals from the screen edge and keeps your sessions alive.

CPU-first, eager-execution tensor/autograd runtime and LLM inference engine written in pure Zig

Native AI workspace for macOS. Local and hosted models, agents, browser, terminal, voice, and mobile access. Locus is wallet-free; LocusX ad

On-device streaming speech-to-text for macOS, built on Qwen3-ASR via MLX

Keep your MacBook working with the lid closed — screen dark, screen locked, everything still running.

Open-source macOS display manager. DDC brightness/contrast/volume, resolution switching, window tiling, night shift, HDR, profiles, CLI. Fre

Which local LLMs can your computer run? Hardware-aware CLI, TUI and GUI: yes/slow/no verdicts and tok/s from your GPU VRAM and RAM, then ver

Native macOS image generator for FLUX.2 Klein — fully on-device on Apple Silicon via MLX. No Python, no ComfyUI, no server.

Local-first meeting recorder for Mac. Whisper on-device, your AI for summaries, MCP server for Claude Desktop.

Lightweight, open-source macOS menu bar system monitor — CPU, memory, network, disk, GPU, temperatures, battery. No telemetry, Apple Silicon

Thermal guard for Apple Silicon: pauses hot jobs with SIGSTOP and resumes them cool, optional admission queue against a thermal budget, keep

Local-first macOS AI workspace. Notch-native, 8 providers, your coding agents and their subagents in one place. ⌥Space anywhere.

Run models too big for RAM on Apple Silicon — any Hugging Face model, dense or MoE — by streaming weights from disk

Catch local LLMs spilling into the CPU. One run shows GPU placement, prefill, decode, memory, power and a verdict.

macOS menu bar app that translates selected text in place with a hotkey: up to three language pairs with automatic direction, your writing s

Native Rust, single-binary MLX inference server for Apple Silicon. OpenAI/Anthropic-compatible LLM serving — text, vision, audio, embeddings

Local AI model manager

Run stock llama.cpp with the stock Unsloth Qwen3.6-35B-A3B 4-bit on a 16 GB Mac.

Run Claude Code with OpenRouter models while keeping tools, file editing, bash, MCP, VS Code support, smart caching, model switching, and sa

IPTV player for Mac, Windows, Android and Android TV: Xtream Codes and M3U, instant search, built-in mpv player, subtitles, Picture in Pictu

Free Mac screen recorder and screenshot tool for polished software demos: automatic zoom, smooth cursor, captions, camera, and a built-in MC

Typed decisions (choice / score / yes-no) from local Qwen models on Apple Silicon. Probabilities come straight from the logits, no text gene

A docker compose for Apple's native container runtime — run your compose.yaml on Apple silicon, no Docker daemon.

Krea 2 Turbo (12.9B text-to-image) ported to Apple MLX — local web UI + CLI, validated faithful to PyTorch

Local workload scheduler for autonomous agents sharing an Apple Silicon Mac.

A phone OS written from scratch by Claude: Slab UI, own compositor, DHCP/DNS/HTTP stack, HTML/CSS browser and a YouTube player. Runs in QEMU

Tiny engine, immense models — run large MoE LLMs (gpt-oss, Mixtral, Qwen3-MoE) on ordinary machines by streaming experts from disk. OpenAI-c

macOS-Port von kimi-k3-in-c: Patch, Installer und Validierung für Apple Silicon. Ohne Modellgewichte.