Update app.py

app.py

@@ -10,7 +10,6 @@ if not hf_token:
 # Initialize InferenceClient
 client = InferenceClient(model="huihui-ai/Llama-3.3-70B-Instruct-abliterated", token=hf_token)
 
-
 def respond(
     message,
     history: list[tuple[str, str]],
@@ -33,9 +32,10 @@ def respond(
     response = ""
 
     try:
+        # Use 'messages' instead of 'inputs'
         for message in client.chat_completion(
             model="huihui-ai/Llama-3.3-70B-Instruct-abliterated",
-            inputs=messages,
+            messages=messages,  # Correct argument
             parameters={
                 "max_tokens": max_tokens,
                 "temperature": temperature,
@@ -47,10 +47,8 @@ def respond(
             response += token
             yield response
     except Exception as e:
-        yield f"Error: {str(e)}"
-
+        yield f"Error: {str(e)}" 
 
-# Gradio Chat Interface
 demo = gr.ChatInterface(
     respond,
     additional_inputs=[