adding some llm configurations

app.py

@@ -18,6 +18,10 @@ from langchain.document_loaders import WebBaseLoader
 
 from typing import Iterator
 
+MAX_MAX_NEW_TOKENS = 2048
+DEFAULT_MAX_NEW_TOKENS = 1024
+MAX_INPUT_TOKEN_LENGTH = int(os.getenv("MAX_INPUT_TOKEN_LENGTH", "4096"))
+
 
 text_splitter = RecursiveCharacterTextSplitter(chunk_size=350, chunk_overlap=10)
 model_id = HuggingFaceHub(repo_id="HuggingFaceH4/zephyr-7b-beta", model_kwargs={"temperature":0.1, "max_new_tokens":300})