Multimedia Agent: Agentic Multimedia File Process using Aulti Agent cooperation

This is a Python demonstration of a multi agent cooperate for multimedia file process. The main steps are:

1. Prompt an LLM to triage agent;
2. Triage agent transfer the conversation to that agent;
3. Specific agent calls the tool to obtain the model and then processes it to reply to the user with relevant results

Getting Started

To get started with MultimediaAgent, follow these steps:

Installation:

git clone https://github.com/wuhongsheng/MultimediaAgent.git
pip install -r requirements

Usage:

huggingface-cli download --resume-download openbmb/MiniCPM-V-2_6-gguf ggml-model-Q4_K_M.gguf --local-dir ./models/
huggingface-cli download --resume-download openbmb/MiniCPM-V-2_6-gguf mmproj-model-f16.gguf --local-dir ./models/
export API_KEY="your api key"
export API_BASE="your api server url"
python agents.py

License

Multimedia Agent is released under the MIT License. You are free to use, modify, and distribute the code for both commercial and non-commercial purposes.