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

# Load model and tokenizer
tokenizer = AutoTokenizer.from_pretrained(
    "stabilityai/stable-code-3b", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(
    "stabilityai/stable-code-3b",
    trust_remote_code=True,
    torch_dtype="auto",
).to("cuda" if torch.cuda.is_available() else "cpu")  # Check for GPU availability

# Define the main function for code generation
def generate_code(prompt):
    inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
    tokens = model.generate(
        **inputs,
        max_new_tokens=48,
        temperature=0.2,
        do_sample=True,
    )
    generated_code = tokenizer.decode(tokens[0], skip_special_tokens=True)
    return generated_code

# Define the Gradio interface
iface = gr.Interface(
    fn=generate_code,
    inputs=[gr.Textbox(lines=2, placeholder="Enter your Python code prompt")],
    outputs="textbox",
    title="Python Code Completion",
    description="Generate code completions using a large language model.",
)


# Launch the Gradio app
iface.launch()