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

model_id = "codellama/CodeLlama-7b-Python-hf"

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto", torch_dtype="auto")
pipe = TextGenerationPipeline(model=model, tokenizer=tokenizer)

def generate_code(prompt):
    result = pipe(prompt, max_new_tokens=200, do_sample=True, temperature=0.5)
    return result[0]["generated_text"]

iface = gr.Interface(fn=generate_code, inputs="text", outputs="text")
iface.launch()