Ukrainian software developer MacPaw and foundation model company Liquid AI have announced a strategic, long-term partnership to co-develop the technology stack for local AI on the Mac. The partnership stems from a shared vision. Both companies believe that the everyday intelligence people rely on should be able to run on their personal computers.
The collaboration combines Liquid Foundation Models — which will be designed specifically for macOS AI assistance — with Elix and Mnemos, MacPaw’s own on-device inference and memory technologies.
For Mac users, this provides AI that’s designed to keep personal data private on their Mac, respond quickly, while handling core tasks without an internet connection. Eney, MacPaw’s AI assistant for macOS, will be the first product to benefit from this partnership; results are expected later this year.
Deep macOS Engineering
“After almost two decades of building software for the Mac, MacPaw is evolving its standalone products into a connected ecosystem, with AI as a core technology we build and own,” says Oleksandr Kosovan, CEO and founder of MacPaw.
“We believe intelligence should live where people work: private by design, fast by default, and be able to reach the cloud when that’s the better tool. By combining MacPaw’s engineering expertise with Liquid AI’s model efficiency, we are building the AI stack for the Mac, one where your personal data stays on device and the heavy lifting happens wherever it makes the most sense.”
Liquid AI’s models already bring on-device intelligence to companies like Mercedes-Benz, Insilico Medicine, and Shopify. MacPaw and Liquid AI will jointly build a new local AI architecture for the Mac, with the work running in two directions.
Project’s Aims
Firstly, the cooperation between MacPaw and Liquid AI will bring Eney’s intelligence on device. Liquid AI develops and fine-tunes its foundation models for the assistant’s real tasks, so its core work runs locally through Elix on Apple silicon, while cloud models will remain available where they are the better tool. The project will also develop Mnemos, MacPaw’s memory layer, which lets the assistant retain context across interactions and become more useful over time.
That combination of on-device intelligence, persistent memory and native macOS task execution in a single product has not existed on the Mac before. Running a model locally is one layer and is usually the easy one. The hard parts come next with choosing the right model for each task, building the agentic flows around it and proving those flows can work reliably on local models.
MacPaw and Liquid AI say that their partnership will build the full stack that solves this. Models will be adapted to a real assistant’s tasks, an inference framework built for Apple silicon, with persistent memory and native macOS task execution, integrated in a single product. The two companies say that the combination has not existed on the Mac before.
Efficient Foundation Models
“We build efficient foundation models and the tools around them so that companies can bring intelligence onto the devices their customers already use,” says Ramin Hasani, co-founder and CEO of Liquid AI. “This partnership will bring efficient, private, on-device LFMs to millions of Mac users.”
For MacPaw and Liquid AI, implementing the on-device AI stack in Eney is intended as the starting point of a longer-term plan. The models, inference framework and memory layer are designed as shared infrastructure, built for the MacPaw ecosystem of products and beyond. Once deployed in Eney, the same technologies could eventually be made globally available to thousands of Mac developers through Setapp, MacPaw’s unified software marketplace.
