FunctionGemma

Name: FunctionGemma
Author: Google

GoogleOpen WeightsVerified

FunctionGemma is a specialized version of Gemma 3 270M fine-tuned for function calling and designed to run on edge devices. It bridges natural language and software execution, translating user commands into executable API actions. The model excels at unified action and chat capabilities, switching seamlessly between generating structured function calls and conversational responses. Built specifically for customization through fine-tuning, it demonstrated 85% accuracy on Mobile Actions after training (up from 58% baseline). Small enough to run on mobile phones and edge devices like NVIDIA Jetson Nano, it uses Gemma's 256k vocabulary to efficiently tokenize JSON and multilingual inputs.

2025-12-18

270M

Decoder-only Transformer

Apache-2.0

Compare with other models

Specifications

Parameters: 270M
Architecture: Decoder-only Transformer
License: Apache-2.0
Context Window: 8,192 tokens
Type: text
Modalities: text

Benchmark Scores

view all (+1)

Advanced Specifications

Model Family: Gemma
Finetuned From: Gemma 3 270M
API Access: Available
Chat Interface: Not Available
Multilingual Support: Yes
Variants: FP32FP16INT8GGUFMLX
Hardware Support: NVIDIA Jetson NanoNVIDIA RTX PRONVIDIA DGX SparkMobile phonesEdge devicesCPUCUDAMetalWebGPUTPU

Capabilities & Limitations

Capabilities: function callingtool usestructured outputJSON generationmultilingualoffline inferencefine-tuningedge deploymentmobile actionslow latencyprivacy-first
Known Limitations: Requires fine-tuning for production-grade performanceLimited to defined API surfacesNot suitable for zero-shot prompting in complex scenariosBest for local, deterministic tasks rather than general conversation
Notable Use Cases: Mobile voice assistantsSmart home controlMedia playback controlNavigation commandsCalendar and contact managementOn-device game mechanicsOffline personal assistantsEdge AI agentsCustom API automationLocal-first applications
Function Calling Support: Yes
Tool Use Support: Yes

Resources

Related Models

Claude 4.5 Opus

Anthropic

Claude 4.5 Opus is the most intelligent and capable model in the Claude 4.5 family, designed for the most demanding reasoning, research, and coding tasks. It introduces a unique 'effort' parameter (low, medium, high) that allows users to control the depth of the model's reasoning process and token usage. Opus 4.5 excels in handling ambiguity, complex problem-solving, and deep analysis, often succeeding where other models fail on multi-system bugs or intricate research questions. It is priced to be more accessible than previous Opus generations while delivering state-of-the-art performance.

Gemini 3 Pro

Google

Gemini 3 Pro is Google's flagship multimodal foundation model, released in November 2025. Built on a sparse Mixture-of-Experts (MoE) Transformer architecture, it features a 1 million token context window and native understanding of text, images, audio, and video. The model introduces 'Deep Think' capabilities for enhanced reasoning, controlled via a 'thinking_level' parameter, and is optimized for 'agentic' workflows and 'vibe coding'—the generation of full applications from natural language. It supports advanced function calling and tool use, making it suitable for complex software engineering and long-context analysis tasks.

Typemultimodal

ParametersProprietary

2025-11-18

Proprietary

Details Compare

Claude 4.5 Haiku

Anthropic

Claude 4.5 Haiku is the fastest and most cost-efficient model in the Claude 4.5 family, designed to deliver near-frontier intelligence with low latency. It is optimized for high-volume tasks, real-time responses, and sub-agent orchestration. Despite its smaller size, it supports advanced features like 'extended thinking' and computer use, making it suitable for parallelized workflows where speed is critical. It is positioned as a drop-in replacement for previous mid-tier models with significantly better economics.