On-Device Function Calling in Google AI Edge Gallery

2026-02-25 · Source: Google Developers Blog - AI · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering, Internet of Things (IoT) & Connected Devices · Depth: Intermediate, quick

Key takeaway

For mobile application developers aiming to integrate advanced AI capabilities, you should explore FunctionGemma in the Google AI Edge Gallery. This model allows you to build responsive, "agentic" applications that perform complex tasks offline, significantly improving user experience by reducing latency and ensuring privacy without cloud dependency.

Key insights

FunctionGemma enables efficient, offline AI function calling directly on mobile devices via Google AI Edge.

Principles

• On-device AI enhances privacy and reduces latency.
• Small models can enable complex, agentic behaviors.

Method

FunctionGemma integrates with Google AI Edge and LiteRT-LM to execute predefined functions locally, allowing AI models to interact with device capabilities and external systems offline.

In practice

• Develop offline calendar management features.
• Implement device hardware control via AI.
• Create AI-driven game logic.

Topics

• FunctionGemma
• On-Device Function Calling
• Edge AI
• Mobile AI Development
• LiteRT-LM

Best for: AI Architect, CTO, VP of Engineering/Data, AI Engineer, Machine Learning Engineer, Software Engineer

Related on AIssential

• Day 14: Function Calling Explained Simply (For DevOps & Cloud Engineers) — Function Calling empowers LLMs to interact with external systems, transforming them from text generators into actionable AI agents.
• Google AI Edge Gallery — Google's new app enables on-device execution of Gemma models and tool calling directly on iPhones.
• GPT-5.4 Is Here: OpenAI’s Most Capable Model Yet Redefines Professional AI Work — GPT-5.4 integrates advanced reasoning, autonomous operation, and enterprise efficiency into a single AI system.
• From 46% to 90%: Fine-Tuning Tiny LLMs for On-Device Agents — Cormac Brick, Google — Fine-tuning tiny LLMs with synthetic data dramatically enhances on-device agent performance and reliability for specific tasks.
• Give it tools - Perplexity — Perplexity models integrate built-in and custom function tools to extend capabilities beyond internal knowledge.
• The Death of the “Everything Prompt”: Google’s Move Toward Structured AI — Google's Interactions API provides structured state management and asynchronous agent orchestration for complex AI applications.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Google Developers Blog - AI.