app.py

import gradio as gr
from transformers import AutoTokenizer, AutoModelForCausalLM

def code_v1(input_text):
    # load tokenizer
    tokenizer = AutoTokenizer.from_pretrained('replit/replit-code-v1-3b', trust_remote_code=True)

    # load model
    model = AutoModelForCausalLM.from_pretrained('replit/replit-code-v1-3b')

    # single input encoding + generation
    x = tokenizer.encode(input_text, return_tensors='pt')
    y = model.generate(x)
    
    # decoding, clean_up_tokenization_spaces=False to ensure syntactical correctness
    generated_code = tokenizer.decode(y[0], skip_special_tokens=True, clean_up_tokenization_spaces=False)
    return generated_code

prompt = gr.inputs.Textbox(label="Prompt")
run_button = gr.outputs.Button(label="Run")
output_prompt = gr.outputs.Textbox(label="Output")

iface = gr.Interface(fn=code_v1, inputs=prompt, outputs=output_prompt)
iface.launch()