Update app.py

app.py

@@ -37,10 +37,12 @@ llm = HuggingFaceLLM(
     model_kwargs={"quantization_config": quantization_config},
     generate_kwargs={"temperature": 0.1, "top_k": 50, "top_p": 0.95},
     device_map="cuda:0",
+    token = hf_token,
 )
 
 embed_model = HuggingFaceEmbedding(
     model_name="kheopss/kheops_embedding_e5_v3",
+    token = hf_token,
 )
 Settings.llm=llm
 Settings.embed_model=embed_model