MacPaw AI is our applied-AI direction - initiatives that help Mac apps think, remember, and get things done right on the device.
They bring inference, memory, and agents together in a native Swift runtime built with the aim of speed and privacy.
We are implementing them in our products. Now, we're opening it to yours.
Everyone interacting with MacPaw and its products - developers, product teams, and partners.
Build on-device AI for Mac in Swift. Describe what your app needs to do, and Elix handles the runtime - from execution and resource management to inference with Liquid Foundation Models.
Learn about ElixWrite the intent in Swift. Elix works out what can run in parallel, so you can spend less time wiring AI infrastructure together. The pipelines layer is open source under Apache 2.0.
One service runs AI across apps and monitors resources. When the LLM isn't needed, Elix unloads it and frees the memory.
Our custom MLX-based inference engine, tuned for Liquid Foundation Models and Apple silicon - running fully on the Mac.
Give your AI a memory that stays on the Mac. It keeps context over time, knows where each fact came from, and is built with the aim of privacy by design.
Meet MnemosPeople, projects, and facts stay connected instead of becoming disconnected chunks. When information conflicts, Mnemos can resolve it - without losing track of the source.
Mnemos keeps apart what you said, what it found in its data, and what it observed. Each has its own trust level and traceable source.
Mnemos uses Portable Memory, an open, vendor-neutral format - export your memory, merge it, or move it between apps and devices.
Setapp AI Gateway connects your app to leading AI models through one compatible endpoint. No pile of API keys. No billing system to build.
Explore the gatewayClaude, GPT, Gemini, DeepSeek, and 10+ more are available through the same endpoint. Want another model? Change one parameter.
Smart routing sends simpler requests to lower-cost models, while caching avoids paying twice for repeat work.
Setapp already handles subscriptions and AI credit balance. You can focus on the AI feature instead of building the machinery around it.
See how MacPaw AI works in products people already use.
EneyEney is an AI assistant that helps you get things done on your Mac - without sending every task to the cloud.
Local features plan to run on Elix, with Mnemos providing long-term memory. Our goal is fast, private AI that can work with you in real time.
Visit
CleanMy®Phone EngineThe CleanMy®Phone Engine uses machine learning to make sense of your photo library and clear out what you don't need.
It spots blurry photos, duplicates, and forgotten screenshots, helping you free up space while keeping the photos that matter.
Visit
SetappSetapp gives you one subscription to a curated collection of Mac and iOS apps. You can also buy single-app subscriptions or add an AI+ plan.
Setapp is adding an AI assistant that matches what you want to do with the right app, while Setapp AI brings leading AI models into member apps - all covered by the same subscription.
VisitSome of our AI work is too useful to keep to ourselves. Explore the models and datasets MacPaw Research has made open source.
BLIP-based captioner for macOS icons - drives accessibility labelling across our pipelines.
View on Hugging Face ↗YOLOv11-Large fine-tuned for macOS UI element detection - the backbone behind Screen2AX.
View on Hugging Face ↗Sibling YOLOv11-Large model - detects UI groups like toolbars and tab bars.
View on Hugging Face ↗UI Parsing and Accessibility Dataset - 352 macOS app screens with accessibility trees.
View on Hugging Face ↗Structured metadata for thousands of Mac App Store apps.
View on Hugging Face ↗Automated macOS UI exploration trajectories for GUIrilla-See.
View on Hugging Face ↗Tools, ports, and engines we built for our own AI work and made available on GitHub.
SwiftUI-like DSL for composing AI pipelines in Swift.
View on GitHub ↗Community-driven Swift package for the OpenAI API.
View on GitHub ↗Swift port of Gliner2 - entity recognition on the Mac.
View on GitHub ↗Fast-dLLM diffusion-LLM inference, ported to MLX.
View on GitHub ↗Extract structured accessibility trees from any Mac app.
View on GitHub ↗Automated agent for exploring and testing macOS GUI flows.
View on GitHub ↗An open, vendor-neutral memory format and protocol for AI.
View on GitHub ↗Swift implementation of the portable memory protocol.
View on GitHub ↗Researching how AI can work better on the Mac - from portable memory and computer-use agents to AI pipelines and code evaluation.
We work with specialists across foundation models, voice AI, and model training to bring the right technology into MacPaw AI initiatives.
Liquid AI builds efficient Liquid Foundation Models. Elix is designed to run them even leaner on Apple silicon.
Respeecher brings studio-grade voice cloning and real-time text-to-speech to Mac voice features through Elix.
distil labs helps us train purpose-built small language models that can handle specific tasks without an expensive LLM call.
We work with universities on AI research, internships, labs, and opportunities for students to learn by doing.
Joint research and internships in on-device inference.
Joint applied-AI research and a steady ML talent pipeline.
Joint research, internships, and a talent pipeline.
Home of our "Bilka Space" AI Lab and an ML scholarship.
Working on AI for the Mac? Bring us the product, technical problem, or idea you're exploring. We can talk about platforms, partnerships, model licensing, and roadmap.
Email the team → [email protected]See what the press is saying about our AI products, platform, partnerships, and plans.
