Update app.py

app.py

@@ -17,7 +17,7 @@ model = AutoModelForCausalLM.from_pretrained(
 model.eval()  # Set the model to evaluation mode
 
 # Initialize the inference client (if needed for other API-based tasks)
-client = InferenceClient(provider="together",token=access_token)
+client = InferenceClient(provider="together", token=access_token)
 
 def conversation_predict(input_text):
     """Generate a response for single-turn input using the model."""
@@ -50,31 +50,18 @@ def respond(
 
     messages.append({"role": "user", "content": message})
 
-    # response = ""
-
-    # Stream response tokens from the chat completion API
-    # for message_chunk in client.chat_completion(
-    #     model = "google/gemma-2b-it",
-    #     messages=messages,
-    #     max_tokens=max_tokens,
-    #     stream=True,
-    #     temperature=temperature,
-    #     top_p=top_p,
-    # ):
-    #     token = message_chunk["choices"][0]["delta"].get("content", "")
-    #     response += token
-    #     yield response
-
-    response = client.chat.completions.create(
-        model = "google/gemma-2b-it",
+    # Get the complete response at once (no streaming)
+    response = client.chat_completion(
+        model="google/gemma-2b-it",
         messages=messages,
         max_tokens=max_tokens,
         stream=False,
         temperature=temperature,
         top_p=top_p,
     )
-    response += response.choices[0].message
-    return response
+    
+    # Extract and return the full response
+    return response["choices"][0]["message"]["content"]
 
 # Create a Gradio ChatInterface demo
 demo = gr.ChatInterface(
@@ -94,4 +81,4 @@ demo = gr.ChatInterface(
 )
 
 if __name__ == "__main__":
-    demo.launch(share=True)
+    demo.launch(share=True)