Update app.py

app.py

@@ -11,25 +11,19 @@ client = OpenAI(
     api_key=ACCESS_TOKEN,
 )
 
-@tenacity.retry(wait=tenacity.wait_exponential(multiplier=1, min=4, max=10))
+# Retry logic with tenacity for handling API rate limits
+@tenacity.retry(wait=tenacity.wait_exponential(multiplier=1, min=4, max=10), stop=tenacity.stop_after_attempt(5))
 async def respond(
     message,
-    history,
     system_message,
     max_tokens,
     temperature,
     top_p,
 ):
     try:
-        messages = [{"role": "system", "content": system_message}]
-
-        for val in history:
-            if val[0]:
-                messages.append({"role": "user", "content": val[0]})
-            if val[1]:
-                messages.append({"role": "assistant", "content": val[1]})
-
-        messages.append({"role": "user", "content": message})
+        # Only use the system message and the current message for the response
+        messages = [{"role": "system", "content": system_message},
+                    {"role": "user", "content": message}]
 
         response = ""
         # Properly stream chat completions using dot notation
@@ -41,11 +35,15 @@ async def respond(
             top_p=top_p,
             messages=messages,
         )
-        for chunk in stream:  # Iterate over the streamed response chunks
+
+        # Stream response and concatenate tokens
+        for chunk in stream:
             if hasattr(chunk.choices[0].delta, 'content'):
                 token = chunk.choices[0].delta.content
                 response += token
+
         return response
+
     except APIError as e:
         error_details = e.body
         error_type = error_details.get("type")
@@ -57,20 +55,18 @@ async def respond(
             error_str = f"{error_type}: {error_message} (code: {error_code}, param: {error_param})"
         else:
             error_str = "An error occurred during streaming"
-
         print(f"Error: {error_str}")
         return error_str
+
     except Exception as e:
         print(f"Error: {e}")
         return "Error occurred. Please try again."
 
 
-# Make the Gradio app async
-async def generate_response(message, history, system_message, max_tokens, temperature, top_p):
-    new_history = history + [[message, ""]]
-    response = await respond(message, history, system_message, max_tokens, temperature, top_p)
-    new_history[-1][1] = response
-    return response, new_history
+# Async Gradio function to handle user input and response generation without history
+async def generate_response(message, system_message, max_tokens, temperature, top_p):
+    response = await respond(message, system_message, max_tokens, temperature, top_p)
+    return response
 
 
 def launch_app():
@@ -79,18 +75,17 @@ def launch_app():
         with demo:
             gr.Markdown("# Chatbot")
             message = gr.Textbox(label="Message")
-            history = gr.State([["", ""]])
             system_message = gr.Textbox(label="System message")
             max_tokens = gr.Slider(minimum=1, maximum=2048, value=2048, step=1, label="Max new tokens")
             temperature = gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature")
             top_p = gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-P")
             response = gr.Text(label="Response")
 
-            # Use the async version of generate_response
+            # Use the async version of generate_response without history
             gr.Button("Generate Response").click(
                 generate_response,
-                inputs=[message, history, system_message, max_tokens, temperature, top_p],
-                outputs=[response, history],
+                inputs=[message, system_message, max_tokens, temperature, top_p],
+                outputs=[response],
                 show_progress=False,
             )
         demo.launch(show_error=True)