Update app.py

app.py

@@ -29,18 +29,18 @@ index = VectorStoreIndex.from_vector_store(vector_store, embed_model=embed_model
 # Load the correct tokenizer and LLM
 from transformers import AutoTokenizer
 
-quantization_config = BitsAndBytesConfig(
-    load_in_4bit=True,
-    bnb_4bit_compute_dtype="float16"
-)
+# quantization_config = BitsAndBytesConfig(
+#     load_in_4bit=True,
+#     bnb_4bit_compute_dtype="float16"
+# )
 
 llm = HuggingFaceLLM(
     model_name="microsoft/phi-1_5", # Mały model 1.3B
     tokenizer=AutoTokenizer.from_pretrained("microsoft/phi-1_5"),
-    device_map="auto",
-    model_kwargs={
-        "quantization_config": quantization_config,
-    }
+    device_map="auto"
+    # model_kwargs={
+    #     "quantization_config": quantization_config,
+    # }
 )
 
 #Query Engine