import gradio as gr
from transformers import AutoTokenizer, AutoModelForCausalLM
import torch

# Load DeepSeek model
model_id = "deepseek-ai/deepseek-llm-7b-chat"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype=torch.float16, device_map="auto")


def generate_response(prompt, temperature,top_p,max_new_tokens,repetition_penalty):
    inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
    outputs = model.generate(
        **inputs,
        do_sample=True,
        temperature=temperature,
        top_p=top_p,
        max_new_tokens=max_new_tokens,
        repetition_penalty=repetition_penalty,
    )

    return tokenizer.decode(outputs[0], skip_special_tokens=True)
    # return "Hello " + name + "!!"

demo = gr.Interface(fn=generate_response, 
                    inputs=[
                        gr.Textbox(label="Prompt", lines=6, placeholder="Ask something..."),
                        gr.Slider(0.1, 1.5, value=0.7, step=0.1, label="Temperature"),
                        gr.Slider(0.1, 1.0, value=0.9, step=0.1, label="top_p"),
                        gr.Slider(32.0, 2048.0, value=512.0, step=1.0, label="max_new_tokens"),
                        gr.Slider(1.0, 2.0, value=1.1, step=0.1, label="repetition_penalty"),
                    ], 
                    outputs="text"
                   )
demo.launch()