Matter-0.2-7B Overview

Matter-0.2-7B is a 7 billion parameter language model, a full-finetune of the Mistral 7B 0.2 architecture. Developed by 0-hero, this model was trained using the proprietary Matter 0.2 dataset, which aggregates data from over 35 distinct datasets, encompassing more than 6 billion tokens.

Key Capabilities

• General Assistant Tasks: Optimized for use as a helpful AI assistant, responding to a wide range of prompts.
• Function Calling: Integrates robust support for function calling, allowing the model to interact with external tools and APIs. This feature uses specific <tool_call> and <tool_response> tokens to structure interactions.
• ChatML Format: Utilizes the ChatML prompt format for structured conversations, including system, user, and assistant roles.

Good For

• Tool-Augmented Applications: Ideal for developers building applications that require the LLM to invoke external functions or retrieve real-time information.
• Conversational AI: Suitable for creating intelligent assistants that can understand and respond to user queries while leveraging external capabilities.