using quantized model to try to fix memory issue

app.py

@@ -32,7 +32,8 @@ MAX_INPUT_TOKEN_LENGTH = int(os.getenv("MAX_INPUT_TOKEN_LENGTH", "4096"))
 
 embeddings = HuggingFaceHubEmbeddings()
 
-model_id = "HuggingFaceH4/zephyr-7b-beta"
+model_id = "TheBloke/zephyr-7B-beta-GGUF"
+# model_id = "HuggingFaceH4/zephyr-7b-beta"
 # model_id = "meta-llama/Llama-2-7b-chat-hf"
 
 # model = AutoModelForCausalLM.from_pretrained(