Liquid AI and MacPaw Partner to Develop On-Device AI Models for macOS Assistant Eney

Liquid AI and MacPaw have announced a long-term strategic partnership to develop and fine-tune on-device foundation models that will power Eney, MacPaw’s artificial intelligence assistant for macOS.

Under the agreement, the companies will jointly build AI models designed to run locally on Apple silicon devices, enabling AI processing directly on users’ computers rather than relying on cloud-based infrastructure.

According to the companies, the models will operate using MacPaw’s Elix inference engine and the Mnemos memory layer, technologies intended to optimize on-device inference, memory management and personalization.

The partnership reflects growing industry interest in edge AI, where machine learning models are executed locally on devices to improve privacy, reduce latency and decrease dependence on internet connectivity.

Running foundation models on Apple silicon is expected to take advantage of the chips’ dedicated AI processing capabilities while enabling faster performance and greater data privacy for macOS users.

The collaboration also highlights a broader trend among software developers to integrate generative AI capabilities directly into desktop operating systems through locally executed models.