AI News Feed
Market watch
Companies

MacPaw Launches Eney, a Proactive AI Assistant for Mac and iPhone

MacPaw has launched Eney, a proactive AI assistant for Mac and iPhone that works across apps, files and the system. It combines local and cloud AI models, the Mnemos memory system and writing tools, after development began in 2019.

Eney started as an idea to simplify MacPaw's CleanMyMac through natural-language commands, founder and CEO Oleksandr Kosovan told 9to5Mac. The project changed as underlying AI technology evolved: first with the boom in large language models, then with agentic AI, and more recently as local AI models became powerful enough to handle increasingly complex tasks that until recently required much larger cloud-based models. MacPaw reworked and refined Eney, which now launches for Mac and iPhone.

MacPaw said Eney is built for the Mac. It understands the context of the machine itself, including the apps a user works across, files on disk and meetings on the calendar, and spots what needs attention before the user does. Examples MacPaw gave include a call about to start, a meeting that needs a summary and a disk running out of space. Eney notices these, prepares the next step and puts it in front of the user for approval without requiring them to switch between multiple apps, tools and products.

9to5Mac highlighted Eney Wand as an early feature. Shaking the cursor over text activates Eney Wand, allowing the user to select that text and prompt Eney about it. Eney Wand can also be invoked with the Option + W shortcut, or by double-tapping the Option key to bring up Eney from anywhere. 9to5Mac said the feature addresses a common pain point with AI tools: having to copy and paste content in and out of a chat window. With system-wide integration, a user can point to something on screen and ask Eney to work with it without moving content between apps.

Eney runs on both local and cloud-based AI models. MacPaw worked with Liquid AI to develop custom models optimized for Apple hardware. 9to5Mac visited Liquid AI's San Francisco headquarters and saw a demonstration of an AI assistant running on an older Mercedes-Benz whose hardware was many years old and, in theory, not powerful enough to run modern AI models. Liquid AI has developed models efficient enough to run on existing hardware with limited computing resources, 9to5Mac reported.

Eney also relies on Mnemos, MacPaw's proprietary AI memory system. Mnemos allows Eney to remember information about users and their preferences and use that context to provide more personalized responses and suggestions. It uses a knowledge graph to connect information about people, projects, preferences and events, and it understands how information changes over time. If a user corrects something or a stored fact becomes outdated, Mnemos can track what is currently relevant and update accordingly. For privacy, Mnemos is designed to store and process memories locally on the Mac, and it distinguishes between information explicitly provided by the user, facts retrieved from connected sources and patterns inferred from previous interactions.

Eney is not entirely on-device, however. More demanding tasks still rely on cloud-based models. Mnemos is also still in development, and not all of its capabilities are available through Eney. For now, its integration is limited to managing Eney's own memories rather than independently accessing information stored in other apps. Kosovan confirmed that Eney already supports importing ChatGPT archives, which lets users bring existing context and preferences into the assistant.

MacPaw also included writing assistance. Eney Spell fixes spelling, grammar and wording in Mail, Slack, Notes and the browser, and works similarly to tools such as Grammarly, according to MacPaw. With the launch, Eney works in eight languages.

Editor's Summary

MacPaw has launched Eney, a proactive AI assistant for Mac and iPhone, after developing the concept since 2019. Eney combines system-wide context awareness, the Eney Wand invocation gesture, local and cloud models developed with Liquid AI, and the Mnemos memory system, while some Mnemos features remain in development.