|
VeloxQuant-MLX — Run Longer Conversations on Your Mac
Free, open-source tool that shrinks the memory a local AI model uses on your Mac, so you can run longer conversations or bigger models. Runs entirely on your...
|
85 % |
|
oMLX — LLM inference, optimized for your Mac
A native macOS inference server built on MLX. Paged SSD KV caching drops agent TTFT from 30-90s to under 5s. OpenAI & Anthropic compatible API for Apple Sili...
|
84 % |
|
AFM — local AI infrastructure for Apple Silicon
A Mac-only, OpenAI-compatible Swift server for running MLX and Apple on-device models locally on Apple Silicon.
|
84 % |
|
Rapid-MLX — Up to 4× Faster Than Apple's MLX | Local LLMs on Apple Silicon
Open-source OpenAI- and Anthropic-compatible LLM server for Apple Silicon. Up to 4× faster than mlx-lm, same weights. Reliable tool calling for coding agents.
|
84 % |
|
Pico AI Server - Local LLM Server for Apple Silicon
Run DeepSeek R1, Llama, Gemma & 300+ AI models locally on Apple Silicon M1/M2/M3. Your private AI server alternative to OpenAI, powered by MLX.
|
83 % |
|
Quadpoint AI Studio — frontier models, resident on your Mac
Quadpoint AI Studio runs frontier open-weight models on Apple silicon with a Swift and Metal engine, a single-file .qpx bundle format, and a local OpenAI-com...
|
83 % |
|
CALMVPS|Premium Mac Bare-Metal Node Rental
Dedicated bare-metal Mac Mini M4 nodes for developers and teams — zero virtualization overhead, full Apple Silicon performance, 10-minute delivery, iOS CI/CD...
|
82 % |
|
MLXHub — Local LLMs on Apple Silicon
Run large language models locally on your Mac with Apple MLX. No Python, no cloud, no hassle.
|
82 % |
|
NovaMLX — The blazing fast pure-Swift LLM server for Apple Silicon
Run 50+ model families natively on your Mac. 100% Swift. Zero Python dependencies. OpenAI & Anthropic compatible.
|
82 % |
|
M1K3 for Mac — Your AI. Your Mac. Private by design.
M1K3 is on-device AI for macOS. Local inference, live voice, a knowledge graph, encrypted call memory, and an MCP server — on your Mac by default, private by...
|
81 % |
|
Vesta AI Explorer — AI Chat for macOS
|
81 % |
|
M1K3 for Mac — Your AI. Your Mac. Private by design.
M1K3 is on-device AI for macOS. Local inference, live voice, a knowledge graph, encrypted call memory, and an MCP server — on your Mac by default, private by...
|
81 % |
|
Local AI for Mac: chat, code, images, offline · Slate
Run AI models locally on your Mac. Chat, a coding agent, image generation and dictation in one native app, offline by default. Free and open source, with an...
|
81 % |
|
ToshLLM | Local AI for Intel Macs with AMD GPUs
Run open language models locally on Intel Macs with AMD GPUs. ToshLLM provides native Metal acceleration, private chat, vision and reproducible benchmarks.
|
81 % |
|
Pexper — Free AI Workflow Builder | Any Model, Local or Cloud
Free node-based AI workflow builder. Connect OpenAI, Claude, Stable Diffusion, or local models on a visual canvas. Text, images, video, audio — no code requi...
|
80 % |
|
MacSentinel — Mac System Monitor & Storage Cleaner
Free Mac system monitor and storage cleaner for Apple Silicon & Intel. See live RAM, CPU and disk usage, and reclaim GBs of Xcode, Docker and app caches.
|
80 % |
|
vMLX — Run LLMs Locally on Your Mac | MLX for Apple Silicon
Run LLMs locally on your Mac. An MLX inference engine for Apple Silicon with SSD prefix caching, paged KV cache and continuous batching. Free and fully offline.
|
80 % |
|
MACCOME|Rent a Mac|Cloud Mac & Mac Mini Cloud
Rent a Mac on dedicated M4 bare metal—Mac hosting and Mac mini rental for Xcode, CI, on-device AI, and 4K media. Provision your cloud Mac in minutes.
|
80 % |
|
MetalMind — Local AI engine and agent for Mac
A native macOS app that runs local AI on Apple Silicon. Open-weights LLMs from Hugging Face, offline, with workspace tools, diff-approved file edits, and an...
|
80 % |
|
gpu-io.net — Dedicated Apple Silicon in Stockholm
Bare-metal Mac Studio hosting in Stockholm. M3 Ultra with 512 GB unified memory. Dedicated machines and inference API.
|
80 % |
Detected social media networks