Update app.py

app.py

@@ -11,7 +11,6 @@ import faiss
 from transformers import (
     AutoTokenizer, 
     AutoModelForCausalLM, 
-    BitsAndBytesConfig, 
     AutoModel, 
     TextIteratorStreamer
 )
@@ -384,13 +383,13 @@ else:
     index = faiss.read_index('./storage/faiss_index.faiss')
 
 # Load the model
-nf4_config = BitsAndBytesConfig(
-        load_in_4bit=True,
-        bnb_4bit_quant_type="nf4",
-    )
+# nf4_config = BitsAndBytesConfig(
+#         load_in_4bit=True,
+#         bnb_4bit_quant_type="nf4",
+#     )quantization_config = nf4_config, 
 
 tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b-it", token=HF_TOKEN)
-model = AutoModelForCausalLM.from_pretrained("google/gemma-2b-it", quantization_config = nf4_config, token=HF_TOKEN)
+model = AutoModelForCausalLM.from_pretrained("google/gemma-2b-it", token=HF_TOKEN)
 
 
 def make_inference(query, hist):