from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline

model = AutoModelForCausalLM.from_pretrained("VishalMysore/cookgptlama")
tokenizer = AutoTokenizer.from_pretrained("VishalMysore/cookgptlama")

from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline

model_id = "VishalMysore/cookgptlama"

tokenizer = AutoTokenizer.from_pretrained(model_id)
model_8bit = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto",load_in_8bit=True)
print(model_8bit.get_memory_footprint())

pipe = pipeline("text-generation", model=model, tokenizer=tokenizer)

# We use the tokenizer's chat template to format each message - see https://huggingface.co/docs/transformers/main/en/chat_templating
messages = [
    {
        "role": "system",
        "content": "you are an expert chef in Indian recipe",
    },
    {"role": "user", "content": "give me receipe for paneer butter masala with cook time diet and cusine and instructions"},
]
prompt = pipe.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
outputs = pipe(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])