Update output parameters of generator and retriever LLM

app.py

@@ -35,7 +35,7 @@ def respond(
 
         for message in client.chat_completion(
             messages,
-            max_tokens=2048,
+            max_tokens=6192,
             stream=True,
             temperature=1.0,
             top_p=top_p,

rag.py

@@ -23,7 +23,7 @@ vector_index = os.getenv("VECTOR_INDEX")
 chat_llm = HuggingFaceEndpoint(
     repo_id="meta-llama/Meta-Llama-3-8B-Instruct",
     task="text-generation",
-    max_new_tokens=6000,
+    max_new_tokens=2048,
     do_sample=False,
 )