IT

Google Unveils Gemma 4 AI Models for On-Device Functionality

Google has launched its new Gemma 4 family of open AI models, designed for efficient operation on smartphones and PCs. This move enables on-device AI capabilities and future support for Gemini Nano 4 across Android.
GL
Aryan Mehta
thegreylens.com

Google has announced the release of its Gemma 4 family of open AI models, a significant development aimed at bringing advanced artificial intelligence capabilities directly to consumer devices. These new models are engineered to run efficiently on a wide range of hardware, from smartphones and personal computers to developer workstations, marking a pivotal step towards more accessible and powerful on-device AI. The primary goal behind Gemma 4 is to empower developers to create AI applications that can function seamlessly both locally and offline, reducing the reliance on constant cloud connectivity.

This latest offering from Google includes four distinct configurations: Effective 2B (E2B), Effective 4B (E4B), 26B Mixture of Experts (MoE), and 31B Dense. The larger 31B and 26B models are optimized for delivering superior performance relative to their parameter count. Google has stated that these models are built upon the same foundational research as its Gemini 3 models and are positioned to complement its existing proprietary AI offerings. The E2B and E4B variants are particularly tailored for mobile and edge devices, including Android smartphones and systems like the Raspberry Pi, promising enhanced user experiences through localized AI processing.

Further bolstering this initiative, Google also provided details on Gemini Nano 4 for Android, which is built upon the Gemma 4 architecture. This integration is expected to bring advanced AI features to Android devices later this year. Applications developed using Gemma 4 today are designed to be compatible with future Gemini Nano 4-enabled devices, with Google working on performance optimizations to improve efficiency and facilitate large-scale deployment within the Android ecosystem. Developers can gain early access to these capabilities through the AICore Developer Preview.

The introduction of Gemma 4 signifies Google's commitment to democratizing AI, enabling developers to innovate on consumer hardware and unlock new possibilities for intelligent applications. By focusing on on-device processing, Google aims to enhance user privacy, reduce latency, and provide a more robust AI experience that is not solely dependent on network availability. This strategic move is poised to accelerate the development of smarter, more intuitive technologies across the global hardware landscape.

---

⚠️ This article used AI assistance. Please verify facts independently.

This article was researched and written with AI assistance based on publicly available news sources. All content is reviewed for accuracy by The GreyLens editorial team. For corrections or feedback: news@thegreylens.com

← Back to News