Update app.py

app.py

@@ -63,11 +63,11 @@ from langchain_huggingface import HuggingFaceEndpoint, ChatHuggingFace
 # 1. Use a model that has high availability in the 2026 Provider network
 # Llama-3.1-8B-Instruct is currently a 'gold standard' for this router
 # repo_id = "meta-llama/Llama-3.1-8B-Instruct"
-# repo_id = "Qwen/Qwen2.5-7B-Instruct"
+repo_id = "Qwen/Qwen2.5-7B-Instruct"
 # repo_id = "google/gemma-2-2b-it"
 # repo_id = "mistralai/Mistral-7B-Instruct-v0.3"
 # repo_id = "HuggingFaceTB/SmolLM2-1.7B-Instruct"
-repo_id = "microsoft/Phi-3.5-mini-instruct"
+# repo_id = "microsoft/Phi-3.5-mini-instruct"
 
 
 llm = HuggingFaceEndpoint(
@@ -75,6 +75,7 @@ llm = HuggingFaceEndpoint(
     task="text-generation",
     max_new_tokens=512,
     huggingfacehub_api_token=os.getenv("HF_TOKEN"),
+    provider = 'together';
     model_kwargs={"wait_for_model": True}
 )