app.py

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

# Load the smaller model and tokenizer
model_name = "distilgpt2"  # A smaller model that should work with 16GB of RAM
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)

# Set the device to GPU if available, else use CPU
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
model.to(device)

def generate_response(prompt):
    # Encode the input prompt
    inputs = tokenizer.encode(prompt, return_tensors="pt").to(device)
    
    # Generate the output sequence
    outputs = model.generate(inputs, max_length=150, num_return_sequences=1, pad_token_id=tokenizer.eos_token_id)
    
    # Decode the generated sequence
    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
    
    return response

# Set up Gradio interface
iface = gr.Interface(
    fn=generate_response, 
    inputs="text", 
    outputs="text", 
    title="Crypto Analysis Model", 
    description="Enter your prompt related to Bitcoin or cryptocurrency."
)

# Launch the interface
iface.launch()