pip install accelerate
from transformers import AutoTokenizer, AutoModelForCausalLM
import transformers
import torch
import streamlit as st

model = "tiiuae/falcon-7b"

tokenizer = AutoTokenizer.from_pretrained(model)
text_input= st.text_area("Enter some text: ")
    if text_input: 
        sequences = pipeline(
        text_input,
        max_length=200,
        do_sample=True,
        top_k=10,
        num_return_sequences=1,
        eos_token_id=tokenizer.eos_token_id,
    )
    for seq in sequences:
        print(f"Result: {seq['generated_text']}")
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    tokenizer=tokenizer,
    torch_dtype=torch.bfloat16,
    trust_remote_code=True,
    device_map="auto",
)