Derived from PANews, Google has released FunctionGemma, a fine-tuned version of Gemma 270M designed for function calling in local or offline agent scenarios. The model supports unified chat and tool execution, custom fine-tuning (with Mobile Actions accuracy improved from 58% to 85%), and is optimized for edge devices like NVIDIA Jetson Nano and smartphones. Google provides download options via Hugging Face and Kaggle, along with fine-tuning and deployment guides for multiple frameworks and tools.
Google Launches FunctionGemma 270M Edge AI Model for Local/Offline Agent Use
KuCoinFlashShare
Summary
Google has launched FunctionGemma, a 270M parameter model optimized for local or offline agent use. The model, based on Gemma, supports unified chat and tool execution, with Mobile Actions accuracy boosted to 85%. It runs efficiently on edge devices like NVIDIA Jetson Nano and smartphones. Google offers downloads via Hugging Face and Kaggle, along with guides for fine-tuning and deployment. As liquidity and crypto markets evolve, such tools may help in Countering the Financing of Terrorism through enhanced transaction monitoring.
Source:Show original
Disclaimer: The information on this page may have been obtained from third parties and does not necessarily reflect the views or opinions of KuCoin. This content is provided for general informational purposes only, without any representation or warranty of any kind, nor shall it be construed as financial or investment advice. KuCoin shall not be liable for any errors or omissions, or for any outcomes resulting from the use of this information.
Investments in digital assets can be risky. Please carefully evaluate the risks of a product and your risk tolerance based on your own financial circumstances. For more information, please refer to our Terms of Use and Risk Disclosure.