from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

MODEL_NAME = "deepseek-ai/deepseek-llm-67b-chat"

def load_model():
    tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
    model = AutoModelForCausalLM.from_pretrained(
        MODEL_NAME, torch_dtype=torch.float16, device_map="auto"
    )
    return tokenizer, model

tokenizer, model = load_model()

def generate_response(prompt):
    inputs = tokenizer(prompt, return_tensors="pt").to("cuda")
    outputs = model.generate(**inputs, max_length=500)
    return tokenizer.decode(outputs[0], skip_special_tokens=True)