Gemini Spark, Google’s agentic assistant, is now available on Mac
Google has launched Gemini Spark, its agentic AI assistant, for macOS, enhancing its competitive stance against rivals like Microsoft Copilot and Claude Desktop. This new release integrates with various apps, offering real-time topic tracking and expanded functionalities for everyday digital tasks.
Google has unveiled Gemini Spark for macOS, its agentic AI assistant designed to streamline various aspects of digital life. This launch positions Spark to better compete with existing desktop AI agents by enabling it to interact with local files and, eventually, handle remote tasks directly from the user’s computer.
The new macOS version of Spark is integrated into the existing Gemini desktop app and introduces several enhancements. These include the ability to stay updated on topics in real time and connect with additional Google services like Tasks and Keep, addressing previous user feedback about missing integrations. Furthermore, Spark now supports third-party applications such as Canva, Dropbox, Instacart, OpenTable, and Zillow Rentals, expanding its utility to cover a wider array of daily activities from reserving tables to managing finances.
Looking ahead, Google plans to introduce multi-step task assignments for Spark on mobile devices, allowing users to leverage the desktop agent for complex operations like extracting information from Mac files. Additionally, the forthcoming support for a custom Model Context Protocol will enable direct integration of favorite apps into Spark, offering a more personalized and tailored assistant experience. Gemini Spark for macOS is currently available in beta to Google AI Ultra subscribers in the U.S.
Related articles
Lumo, Proton’s privacy-focused AI chatbot, gets an upgrade
Proton has upgraded its privacy-focused AI chatbot, Lumo, to version 2.0, introducing advanced features like image recognition and generation, alongside significant speed improvements. This update enhances Lumo’s capabilities for project management and introduces persistent memory, all while maintaining Proton’s strong privacy commitments.
Featuring Every Eval Ever Results on Hugging Face Model Pages
Hugging Face now integrates "Every Eval Ever" (EVE) results directly onto model pages, enhancing transparency and aiding developers in selecting performant models. This feature, visible via a dedicated "Evals" tab, simplifies access to comprehensive LLM evaluation data.
Core dump epidemiology: fixing an 18-year-old bug
OpenAI debugged elusive C++ crashes in its Rockset service using a population-level analysis, revealing an 18-year-old bug in GNU libunwind and a silent hardware corruption issue. This epidemiological approach to debugging allowed the team to identify and fix these critical issues impacting their data infrastructure.