| import argparse | |
| import gradio as gr | |
| from ui import chat | |
| import os | |
| from dotenv import load_dotenv | |
| load_dotenv() | |
| USERNAME = os.getenv("USERNAME") | |
| PWD = os.getenv("PWD") | |
| def main(args): | |
| demo = gr.ChatInterface( | |
| fn=chat, | |
| examples=["Explain the enteerprise adoption challenges", "How can we identify a fraud transaction?", "Por que os grandes modelos de linguagem de AI halucinam?"], | |
| title="Chat and LLM server in the same application", | |
| description="This space is a template that we can duplicate for your own usage. " | |
| "This space let you build LLM powered idea on top of [Gradio](https://www.gradio.app/) " | |
| "and open LLM served locally by [TGI(Text Generation Inference)](https://huggingface.co/docs/text-generation-inference/en/index). " | |
| "Below is a placeholder Gradio ChatInterface for you to try out Mistral-7B backed by the power of TGI's efficiency. \n\n" | |
| "To use this space for your own usecase, follow the simple steps below:\n" | |
| "1. Duplicate this space. \n" | |
| "2. Set which LLM you wish to use (i.e. mistralai/Mistral-7B-Instruct-v0.2). \n" | |
| "3. Inside app/main.py write Gradio application. \n", | |
| multimodal=False, | |
| theme='sudeepshouche/minimalist', | |
| ) | |
| demo.queue( | |
| default_concurrency_limit=20, | |
| max_size=256 | |
| ).launch(auth=(USERNAME, PWD), server_name="0.0.0.0", server_port=args.port) | |
| if __name__ == "__main__": | |
| parser = argparse.ArgumentParser(description="A MAGIC example by ConceptaTech") | |
| parser.add_argument("--port", type=int, default=7860, help="Port to expose Gradio app") | |
| args = parser.parse_args() | |
| main(args) | |