from transformers import AutoTokenizer, AutoModelForCausalLM
from peft import PeftModel
import torch

# Load base model and tokenizer
tokenizer = AutoTokenizer.from_pretrained("meta-llama/Llama-3.2-3B")
base_model = AutoModelForCausalLM.from_pretrained(
    "meta-llama/Llama-3.2-3B",
    torch_dtype=torch.float16,
    device_map="auto"
)

# Load LoRA adapter
model = PeftModel.from_pretrained(base_model, "Abdullah-abushammala/insurance-expert-llama-3b-lora")
model.eval()

def ask_insurance_expert(question):
    prompt = f"Question: {question}\\nAnswer:"
    inputs = tokenizer(prompt, return_tensors='pt', padding=True)
    
    with torch.no_grad():
        outputs = model.generate(
            **inputs,
            max_length=120,
            temperature=0.4,
            do_sample=True,
            top_p=0.8,
            repetition_penalty=1.3,
            no_repeat_ngram_size=3,
            pad_token_id=tokenizer.eos_token_id,
            eos_token_id=tokenizer.eos_token_id
        )
    
    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
    return response.split("Answer:", 1)[1].strip()

# Example usage
answer = ask_insurance_expert("What is a deductible in health insurance?")
print(answer)