Intelligence
  • Secure
  • On-device
  • Mac-native

Mac-native intelligence

MacPaw AI gives Mac products the on-device runtime to think, remember, and act.
Native-Swift inference, memory, and agents — fast and private by design.
Already in our products. Opening to yours.

Explore AI solutions Become a partner
In products like
  • Eney
  • CleanMy®Phone Engine
  • Refolder
Built for

Product teams, R&D groups, and ecosystem partners.

Cloud AI for Mac apps

Setapp AI Gateway

Every top AI model in one place — one OpenAI-compatible endpoint for your apps, with no keys to manage and no billing to build.

Visit Setapp AI Gateway

  • Every model, one endpoint

    Claude, GPT-4o, Gemini, DeepSeek, and 10+ more — switch models by changing a single parameter.

  • Smart routing & caching

    Dynamic routing sends simple queries to cheaper models, and repeats are cached — lower cost, same UX.

  • No billing to build

    Setapp validates subscriptions and AI-credit balances for you — ship AI features without a payment stack.

In development Apple-Native Inference Engine

Elix

A Mac-native on-device inference engine — built on MLX and tuned by MacPaw Research for speed and privacy.

Learn about Elix

  • MLX-based, tuned by MacPaw Research

    Speculative decoding, KV-cache reuse, and quantization on MLX — peak tokens per second.

  • Partner-tuned models + Adapter Store

    Foundation models from strategic partners, plus a growing Adapter Store of LoRA adapters.

  • Composable pipelines

    A Swift-first DSL compiles models, adapters, and tools into one parallelized graph.

In development Long-term memory for agents

Mnemos

A local-first memory layer for AI agents on the Mac — every event becomes part of a living, bi-temporal knowledge base, all on-device.

Learn about Mnemos

  • Seven kinds of memory

    Working, Core, Episodic, Semantic, Procedural, Resource, and Vault — modeled for what each holds, not one vector blob.

  • Hybrid retrieval & graph

    BM25 and dense vectors, fused and reranked, over a bi-temporal graph where facts carry validity windows.

  • Local-first & encrypted

    One libSQL file on-device; secrets in a CryptoKit vault behind Touch ID. Nothing leaves the Mac unless you route it out.

AI in our products

The shipping side of MacPaw AI — spanning on-device engines and cloud-scale platforms.

Browse all on MacPaw.com ↗

Models & datasets

MacPaw Research open-sources models, datasets, and tooling as they mature. A selection of what is shipping today.

Browse all on Hugging Face ↗

Open source

Tools, ports, and engines MacPaw publishes for the community. Built in the open, on GitHub.

Browse all on GitHub ↗

Publications & research

Contributing to the field, not racing it. Selected AI research papers from MacPaw Research.

Browse all on MacPaw Research ↗

Strategic partners

We collaborate, we don't compete. From foundation models to academic research — partnerships are how MacPaw AI ships faster than it could alone.

Academic & research

University collaborations driving co-authored work, internships, and our long-term talent pipeline.

Talk to the team

Build with MacPaw AI

Working on a Mac product that needs on-device intelligence? Want to partner, license a model, or talk roadmap? Drop us a line — we read every email.

Email the team → [email protected]

Press about MacPaw AI

Selected coverage of the AI strategy, partnerships, and product launches.

Curious about what else we’re exploring?

Visit MacPaw Research ↗