Update app.py

app.py

@@ -107,14 +107,15 @@ retriever = db.as_retriever(
 #     streaming=True
 # )
 
-repo_id = "meta-llama/Llama-3.3-70B-Instruct"
-
-llm = HuggingFaceEndpoint(
-    repo_id=repo_id,
-    max_new_tokens=8192,
-    temperature=0.1,
-    huggingfacehub_api_token=HF_TOKEN,
-    streaming=True,
+llm = ChatOpenAI(
+    model="meta-llama/Llama-3.3-70B-Instruct",
+    temperature=0,
+    max_tokens=None,
+    timeout=None,
+    max_retries=2,
+    api_key=HF_TOKEN,  # if you prefer to pass api key in directly instaed of using env vars
+    base_url="https://api-inference.huggingface.co/v1/",
+    stream=True,
     
 )