import gradio as gr from transformers import pipeline, AutoTokenizer, AutoModelForCausalLM model_name = "TinyLlama/TinyLlama-1.1B-Chat-v1.0" tokenizer = AutoTokenizer.from_pretrained(model_name) model = AutoModelForCausalLM.from_pretrained(model_name) pipe = pipeline("text-generation", model=model, tokenizer=tokenizer) def generate_reply(instruction): prompt = f"### Instruction:\n{instruction}\n\n### Response:\n" result = pipe(prompt, max_new_tokens=200)[0]["generated_text"] return result.split("### Response:\n")[-1].strip() gr.Interface( fn=generate_reply, inputs=gr.Textbox(lines=4, placeholder="Enter your question…"), outputs="textbox", title="Customer Support Chatbot", description="Powered by TinyLlama fine-tuned on customer queries" ).launch()