from transformers import AutoModelForCausalLM, AutoTokenizer import torch # ✅ Choose a Hugging Face Model MODEL_NAME = "mistralai/Mistral-7B-Instruct-v0.1" # Change model if needed # ✅ Load Tokenizer & Model tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME) model = AutoModelForCausalLM.from_pretrained( MODEL_NAME, torch_dtype=torch.float16, # Use float16 for efficiency device_map="auto" # Auto-detect GPU/CPU ) def ask_book(question): """Generates a response based on a given book-related question.""" inputs = tokenizer(question, return_tensors="pt").to(model.device) outputs = model.generate(**inputs, max_length=200) response = tokenizer.decode(outputs[0], skip_special_tokens=True) return response # Example test if __name__ == "__main__": print(ask_book("What is the summary of the book?"))