app.py

import gradio as gr

def generate(prompt, max_new_tokens=80, temperature=0.7):
    inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
    output = model.generate(
        **inputs,
        max_new_tokens=max_new_tokens,
        temperature=temperature,
        do_sample=True
    )
    return tokenizer.decode(output[0], skip_special_tokens=True)

iface = gr.Interface(
    fn=generate,
    inputs=[
        gr.Textbox(label="Prompt"),
        gr.Slider(1, 500, value=80, label="Max New Tokens"),
        gr.Slider(0, 1, value=0.7, label="Temperature")
    ],
    outputs="text"
)

iface.launch()