FunctionGemma is a new, lightweight version of the Gemma 3 270M model, fine-tuned to translate natural language into structured function and API calls, enabling AI agents to “do more than just talk” and act.
Launched a few months ago, Gemma 3 270M has evolved into FunctionGemma by gaining native function call capabilities in response to developer demand.
Running locally allows the model to function either as an independent agent for private, offline tasks or as an “intelligent traffic controller” that routes more complex requests to larger, remote models.
This is particularly compelling on-device, where agents can automate complex, multi-step workflows, from setting reminders to toggling system settings. To enable this at the edge, models must be lightweight enough to run locally and specialized enough to be reliable.
Rather than being intended for zero-shot prompting, FunctionGemma is designed to be further customized into fast, private, on-device agents capable of translating natural language into executable API actions, Google explains. This approach is key to achieving production-ready level performance.
In our “Mobile Actions” evaluation, fine-tuning transformed the model’s reliability, boosting accuracy from a 58% baseline to 85%.
The model is engineered to run efficiently on resource-constrained devices like mobile phones and NVIDIA Jetson Nano. It uses Gemma’s 256k vocabulary to tokenize JSON and multilingual inputs efficiently.
FunctionGemma supports what Google calls “unified action and chat”, enabling it to generate structured code/function calls to execute tools and then switch back to natural language to explain the results to the user.
Google also notes that FunctionGemma offers broad ecosystem support, allowing fine-tuning with frameworks such as Hugging Face Transformers, Unsloth, Keras or NVIDIA NeMo and deployment via LiteRT-LM, vLLM, MLX, Llama.cpp, Ollama, Vertex AI or LM Studio.
Google clearly outlines the conditions under which FunctionGemma is the preferred option, including having a defined API surface, being willing to fine-tune it, prioritizing local-first deployment, or building a complex system that can combine on-device and remote tasks.
To showcase the model in action, Google has published several demos, including Mobile Actions, TinyGarden, and Physics Playground, all accessible through the Google AI Edge Gallery app available on the Play Store.
Mobile Actions parses natural language commands like “Create a calendar event for lunch tomorrow,”, “Add John to my contacts”, or Turn on the flashlight”, and maps them to corresponding OS-level tool calls.
TinyGarden is a voice-controlled game where players give natural language commands like “Plant sunflowers in the top row and water them”. The model breaks that down into specific function calls like plantCrop and waterCrop with coordinate targets.
Physics Playground is an interactive physics puzzle demo that uses natural language instructions to control simulation actions within the game. This demos uses Transformer.js to showcase client-side JavaScript integration.
FunctionGemma is available on Hugging Face or Kaggle. Google also provides a Colab notebooks and a mobile-actions dataset to help developers specialize the model
