Arcee Agent

Introduction

Arcee Agent is a state-of-the-art 7B parameter language model designed for efficient function calling and tool usage. It provides capabilities comparable to larger models while maintaining speed and efficiency. Ideal for developers, researchers, and businesses, it enables sophisticated AI-driven solutions with reduced computational requirements.

Architecture

Arcee Agent's architecture is built upon the Qwen2-7B model. It supports advanced function calling and multiple tool use formats such as Glaive FC v2, Salesforce, and Agent-FLAN. The model operates in dual-mode, functioning as both a tool router and a standalone chat agent, providing versatility in various applications.

Training

The model was trained using Spectrum with compute resources provided by CrusoeAI. It excels in executing and chaining function calls, interacting with APIs, performing database operations, generating code, and executing multi-step tasks. Arcee Agent has been benchmarked to exhibit competitive performance against much larger models.

Guide: Running Locally

To run Arcee Agent locally, follow these steps:

1. Environment Setup: Ensure you have Python and necessary libraries installed.
2. Download the Model: Clone the repository from Hugging Face or download the model files directly.
3. Install Dependencies: Use pip or conda to install dependencies listed in the requirements.txt file.
4. Load the Model: Use Hugging Face's Transformers library to load the model.
5. Inference: Run inference scripts to test model capabilities.

For optimal performance, consider using cloud GPUs such as AWS EC2, Google Cloud, or Azure.

License

Arcee Agent is licensed under the Apache-2.0 License. This allows for flexible use and distribution, with the requirement to attribute the original authors.