# Accelerating Gemini Nano models on Pixel with frozen Multi-Token Prediction

DevFeed: [Accelerating Gemini Nano models on Pixel with frozen Multi-Token Prediction](<https://devfeed.tech/articles/accelerating-gemini-nano-models-on-pixel-with-frozen-multi-token-prediction-6744.md>)

Original publisher: [Read original article](<https://research.google/blog/accelerating-gemini-nano-models-on-pixel-with-frozen-multi-token-prediction/>)

Published: 2026-06-26T18:30:00Z

Content type: article

Language: en

Sources: [The latest research from Google](<https://devfeed.tech/sources/the-latest-research-from-google.md>)

Topics: [AI Inference](<https://devfeed.tech/topics/ai-inference.md>), [On-device AI](<https://devfeed.tech/topics/on-device-ai.md>), [Google](<https://devfeed.tech/topics/google.md>), [LLMs](<https://devfeed.tech/topics/llms.md>), [gemma](<https://devfeed.tech/topics/gemma.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [efficiency](<https://devfeed.tech/tags/efficiency.md>), [energy](<https://devfeed.tech/tags/energy.md>), [gemini](<https://devfeed.tech/tags/gemini.md>), [inference](<https://devfeed.tech/tags/inference.md>), [machine-intelligence](<https://devfeed.tech/tags/machine-intelligence.md>), [mobile-systems](<https://devfeed.tech/tags/mobile-systems.md>), [natural-language-processing](<https://devfeed.tech/tags/natural-language-processing.md>), [on-device](<https://devfeed.tech/tags/on-device.md>), [on-device-ai](<https://devfeed.tech/tags/on-device-ai.md>), [phones](<https://devfeed.tech/tags/phones.md>)

## AI overview

Google Research describes a method for retrofitting Multi-Token Prediction onto frozen Gemini Nano v3 production models to accelerate on-device inference on Pixel phones. The approach targets mobile energy and memory constraints, improving the speed and energy efficiency of features such as notification summaries and proofreading without requiring separate drafting models.

## Source excerpt

Machine Intelligence