Update app.py

app.py

@@ -28,14 +28,14 @@ def choose_model(model_name):
         model = "meta-llama/Llama-3.1-8B-Instruct"
 
     client = InferenceClient(model, token=os.getenv('deepseekv2'))
-    return 
+    return client
     
 
 @spaces.GPU(duration=1)
 def respond(message, history: list[tuple[str, str]], model, system_message, max_tokens, temperature, top_p):
 
     print(model)
-    choose_model(model)
+    client = choose_model(model)
     messages = [{"role": "system", "content": system_message}]
 
     for val in history: