app.py

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer
import gradio as gr
import spaces
# Load the model and tokenizer
model_name = "NoaiGPT/merged-llama3-8b-instruct-1720894657"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)

# Move model to GPU if available
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
model.to(device)

# Define the prediction function
@spaces.GPU
def generate_text(prompt):
    # Tokenize the input and move to GPU if available
    inputs = tokenizer(prompt, return_tensors="pt").to(device)
    # Generate text using the model
    outputs = model.generate(inputs.input_ids, max_length=200, num_return_sequences=1)
    # Decode the generated text
    generated_text = tokenizer.decode(outputs[0], skip_special_tokens=True)
    return generated_text

# Define the Gradio interface
interface = gr.Interface(
    fn=generate_text,
    inputs=gr.Textbox(lines=2, placeholder="Enter your prompt here..."),
    outputs="text",
    title="LLaMA 3 Text Generation",
    description="Generate text using the LLaMA 3 model fine-tuned for instruction-following tasks."
)

# Launch the interface
interface.launch()