Loading the model without quantization

app.py

@@ -31,11 +31,10 @@ similarity_retriever = vector_store.as_retriever(
 )
 
 # Load the LLM
-quantization_config = BitsAndBytesConfig(load_in_8bit=True, device = "cpu")
+# quantization_config = BitsAndBytesConfig(load_in_8bit=True, device = "cpu")
 tokenizer = AutoTokenizer.from_pretrained("google/gemma-2-9b-it")
 llm_model = AutoModelForCausalLM.from_pretrained(
-    "google/gemma-2-9b-it",
-    quantization_config=quantization_config,
+    "google/gemma-2-9b-it"
 )
 text_generation_pipeline = pipeline(
     model=llm_model,