update

app.py

@@ -34,9 +34,14 @@ index.add(embeddings)
 
 
 # Step 3: Load the Language Model
-model_name = "HuggingFaceH4/zephyr-7b-alpha"
-tokenizer = AutoTokenizer.from_pretrained(model_name)
-model = AutoModelForCausalLM.from_pretrained(model_name)
+# model_name = "HuggingFaceH4/zephyr-7b-alpha"
+# tokenizer = AutoTokenizer.from_pretrained(model_name)
+# model = AutoModelForCausalLM.from_pretrained(model_name)
+
+model_name = "TheBloke/zephyr-7B-beta-GPTQ"
+tokenizer = AutoTokenizer.from_pretrained(model_name, use_fast=True)
+model = AutoModelForCausalLM.from_pretrained(model_name, device_map="auto", trust_remote_code=False)
+
 
 # Step 4: Define the Retrieval Function
 def retrieve_documents(query, top_k=3):