Update app/main.py

app/main.py

@@ -115,6 +115,7 @@ class HFInferenceLLM(LLM):
     provider: str = "hf-inference"
     api_key: str = ""
     max_tokens: int = 4000  # Increased from 500 to 4000 by default
+    
 
     def __init__(self, model: str, provider: str = "hf-inference", api_key: str = "", max_tokens: int = 4000):
         super().__init__()
@@ -216,7 +217,7 @@ def get_llm(model_name: str):
         try:
             return HFInferenceLLM(
                 model="mistralai/Mistral-7B-Instruct-v0.3",
-                provider="hf-inference",
+                provider="together",
                 api_key=hf_token,
                 max_tokens=4000
             )