Update app.py

app.py

@@ -1,13 +1,22 @@
 import gradio as gr
 import torch
+import bitsandbytes as bnb
 from transformers import AutoTokenizer, AutoModelForCausalLM
 
+# BNB config
+bnb_config = BitsAndBytesConfig(
+    load_in_4bit=True,
+    bnb_4bit_use_double_quant=True,
+    bnb_4bit_quant_type="nf4",
+    bnb_4bit_compute_dtype=torch.bfloat16
+)
+
 # Define the BLOOM model name
 model_name = "CreitinGameplays/bloom-3b-conversational"
 
 # Load tokenizer and model
 tokenizer = AutoTokenizer.from_pretrained(model_name)
-model = AutoModelForCausalLM.from_pretrained(model_name)
+model = AutoModelForCausalLM.from_pretrained(model_name, quantization_config=bnb_config)
 
 def generate_text(user_prompt):
   """Generates text using the BLOOM model from Hugging Face Transformers and removes the user prompt."""