app.py

import spaces
from transformers import pipeline
import gradio as gr
import torch
import logging

logging.basicConfig(level=logging.INFO)
logging.info(f"CUDA available: {torch.cuda.is_available()}, CUDA version: {torch.version.cuda}")

def decomp_create_prompt(input_data: str) -> str:
    before = "# This is the decompiled code:\n"
    after = "\n# What is the source code?\n"
    prompt = before + input_data.strip() + after

    return prompt


pipe = pipeline(model="ejschwartz/decaf-v1-22b-4bit", return_full_text=False)
pipe.model.to("cuda")

@spaces.GPU(duration=180)
def generate(text):
    print(f"Generating text... {text}")
    prompt = decomp_create_prompt(text)
    return pipe(prompt, max_new_tokens=2000)[0]['generated_text']

demo = gr.Interface(fn=generate, inputs="text", outputs="text")
demo.launch()