import os
import gradio as gr
from huggingface_hub import login
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
from transformers import pipeline

# Fetch API token from environment variable
api_token = os.getenv("Llama_Token")

# Authenticate with Hugging Face
login(api_token)

# Load LLaMA 3.2 model and tokenizer with the API token
model_name = "meta-llama/Llama-3.2-1B"
tokenizer = AutoTokenizer.from_pretrained(model_name, token=api_token)
model = AutoModelForCausalLM.from_pretrained(model_name, token=api_token)

# Define the function to generate text
def generate_text(prompt, max_length=100, temperature=0.7):
    inputs = tokenizer(prompt, return_tensors="pt")
    output = model.generate(
        inputs['input_ids'], 
        max_length=max_length, 
        temperature=temperature, 
        pad_token_id=tokenizer.eos_token_id
    )
    return tokenizer.decode(output[0], skip_special_tokens=True)

# Create the Gradio interface
iface = gr.Interface(
    fn=generate_text,
    inputs=[
        gr.Textbox(label="Enter your prompt", placeholder="Start typing...", lines=5),
        gr.Slider(minimum=50, maximum=200, value=100, step=1, label="Max Length"),
        gr.Slider(minimum=0.1, maximum=1.0, value=0.7, step=0.1, label="Temperature"),
    ],
    outputs="text",
    title="LLaMA 3.2 Text Generator",
    description="Generate text using the LLaMA 3.2 model. Adjust the settings and input a prompt to generate responses.",
)

# Launch the Gradio app
iface.launch(share=True)