KishoreK/ActionGemma-9B
ActionGemma-9B is a 9 billion parameter Large Action Model developed by KishoreK, fine-tuned from Gemma2-9B-it. This model combines the multilingual capabilities of Gemma with function calling functionalities derived from the xLAM dataset. It is specifically designed to enable action and function calling across all languages supported by Gemma2, making it suitable for multilingual tool use and API interaction.
Loading preview...
Model Overview
KishoreK/ActionGemma-9B is a 9 billion parameter Large Action Model (LAM) that integrates the multilingual strengths of the Gemma2-9B-it base model with advanced function calling capabilities. It was fine-tuned using the xLAM dataset, which is designed to imbue models with the ability to understand and execute tool calls.
Key Capabilities
- Multilingual Function Calling: The model excels at interpreting user queries and making appropriate function calls in multiple languages, leveraging Gemma2's inherent multilingual support.
- Tool Use Integration: It is specifically trained to handle tool definitions and generate structured function calls, as demonstrated by its ability to parse and respond to API schemas (e.g.,
get_weather,search). - Customizable Chat Template: The model provides a flexible chat template that allows developers to define system instructions, user queries, and available tools, facilitating dynamic interaction with external functions.
Good for
- Building Multilingual AI Assistants: Ideal for creating chatbots or agents that need to perform actions or retrieve information using external APIs across different languages.
- Automating Tasks with Function Calls: Suitable for applications requiring the model to interact with various tools or services based on user intent.
- Research in Large Action Models: Provides a foundation for further exploration and development in the field of models capable of complex action sequencing and tool use.