app.py

import gradio as gr
from transformers import pipeline
import torch
# Initialize the pipeline for text generation
# pipe = pipeline("text-generation", model="cognitivecomputations/dolphin-2.9.4-llama3.1-8b")
pipe = pipeline("text-generation", model="cognitivecomputations/dolphin-2.9.4-llama3.1-8b", torch_dtype=torch.float16)

def respond(
    message,
    history: list[tuple[str, str]],
    system_message,
    max_tokens,
    temperature,
    top_p,
):
    # Prepare conversation history with system message
    conversation_history = system_message + "\n"
    for user_message, assistant_message in history:
        if user_message:
            conversation_history += f"User: {user_message}\n"
        if assistant_message:
            conversation_history += f"Assistant: {assistant_message}\n"
    conversation_history += f"User: {message}\n"

    # Generate response
    response = ""
    result = pipe(
        conversation_history,
        max_length=max_tokens,
        do_sample=True,
        temperature=temperature,
        top_p=top_p
    )[0]["generated_text"]

    # Extract only the new assistant response
    new_response = result.split(conversation_history)[-1].strip()
    for token in new_response:
        response += token
        yield response

# Define Gradio interface
demo = gr.ChatInterface(
    respond,
    additional_inputs=[
        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
        gr.Slider(
            minimum=0.1,
            maximum=1.0,
            value=0.95,
            step=0.05,
            label="Top-p (nucleus sampling)",
        ),
    ],
)

if __name__ == "__main__":
    demo.launch()