app.py

import gradio as gr
from huggingface_hub import InferenceClient
from transformers import AutoTokenizer

# Set the model name and initialize the InferenceClient and tokenizer
model_name = "gpt2"  # Replace with your model's name
client = InferenceClient(model_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)

# Define a function to interact with the model
def chat_with_model(input_text):
    # Tokenize the input text
    inputs = tokenizer(input_text, return_tensors="pt", truncation=True, padding=True, max_length=512)
    
    # Send the request to the Hugging Face Inference API
    result = client.text_generation(
        inputs=inputs["input_ids"].tolist(),  # Send tokenized input
        parameters={"max_length": 150, "temperature": 1.0}
    )
    
    # Decode the generated text back to a readable format
    response = tokenizer.decode(result[0]["generated_text"], skip_special_tokens=True)
    
    return response

# Set up the Gradio interface
interface = gr.Interface(
    fn=chat_with_model,
    inputs=[gr.Textbox(lines=5, placeholder="Enter your text here...", label="Input Text")],
    outputs=gr.Textbox(lines=5, label="Response"),
    title="Hugging Face Chatbot",
    description="A simple chatbot powered by Hugging Face and InferenceClient."
)

# Launch the Gradio app
if __name__ == "__main__":
    interface.launch()