Unlocking the Next Era of On-Device AI with Google Tensor and Pixel

Google unveiled the next generation of on-device AI powered by its custom Tensor SoC and TPU at Google I/O India, debuting Gemma 4 E2B and Functional Gemma models that run natively on Pixel 10 devices for offline privacy. The technology enables AI chat, image recognition, and audio transcription without internet, with applications in retail and automotive industries.

That's a wrap on Google I/O India In a special joint showcase with the Google Pixel team, we officially unveiled how Google Tensor’s custom System-on-Chip SoC and advanced Tensor Processing Unit TPU are driving the next generation of powerful, 100% private on-device AI for from Google Pixel 10 family of devices1 Here is a summary of the groundbreaking Google India I/O Showcases: We debuted Gemma 4 E2B for TPU https://huggingface.co/litert-community/gemma-4-E2B-it-litert-lm/blob/main/gemma-4-E2B-it Google Tensor G5.litertlm —a state-of-the-art, powerful, yet remarkably lightweight model designed to run natively on the Pixel TPU. Gemma 4 ensures your data never leaves the device, guaranteeing absolute offline privacy for edge AI applications. We demonstrated how your on-device TPU can act as a personal travel planner, recipe guide, or home automator completely offline. Additionally, using Functional Gemma https://huggingface.co/litert-community/functiongemma-270m-ft-mobile-actions/blob/main/functiongemma-270m-ft-mobile-actions Google Tensor G5.litertlm running natively on the TPU, users can execute We showcased seamless, internet-free reasoning across text, images, and audio: AI Chat: Instant, deep offline conversations even at 30,000 feet . Ask Image: Snap and identify objects, plants, or issues with 0% internet. Ask Audio: 100% private, on-device audio transcription for lectures and notes. We highlighted how this technology transforms industries right now: Retail: Converting recipe ideas into precise, localized in-store shopping maps completely offline. Automobile: Giving mechanics on-the-spot visual diagnostics from photos of faulty parts. Sign up for the Tensor SDK https://developers.google.com/edge/tensor-sdk , offering the unified developer workflow and This I/O Event was made possible through the collaboration of several teams. We thank them for their significant contributions: Tensor Team for Google I/O India Event: Himangshu Roy, Prakul Sawhney, Malini P V, Chanchal Raj, Chirag Gupta, Akhilesh Ravi, Shubham Saini, Abhishek Jatram, Jui Pradhan, Naina Singla, Jayanthan K, Rishubh Khurana, Sarvesh Das, Alice Zheng, Yishuang Pang, Na Li, Lu wang Pixel Business Team: Lokesh Madala, Amit Roykaran, Mitul Shah, Keya Mukherji, Lokesh Madala, Sneha Budhia https://ai.google.dev/edge/litert/next/tensor ml terms https://ai.google.dev/edge/litert/next/tensor ml terms 1- Devices supported are Pixel 10, Pixel 10 Pro, Pixel 10 Pro XL, Pixel 10 Pro Fold