Update app.py

app.py

@@ -17,12 +17,10 @@ headers = {
 BASE_SYSTEM_MESSAGE = "I carefully provide accurate, factual, thoughtful, nuanced answers and am brilliant at reasoning."
 
 def call_nvidia_api(history, system_message, max_tokens, temperature, top_p):
-    """Chama a API da NVIDIA para gerar uma resposta."""
-    # Prepara as mensagens, incluindo a mensagem do sistema se fornecida
     messages = []
     if system_message:
         messages.append({"role": "system", "content": system_message})
-    messages.extend(history)
+    messages.extend([{"role": "user", "content": msg[0]} for msg in history])
 
     payload = {
         "messages": messages,
@@ -31,13 +29,15 @@ def call_nvidia_api(history, system_message, max_tokens, temperature, top_p):
         "max_tokens": max_tokens,
         "stream": False
     }
+    print("Payload sendo enviado:", json.dumps(payload, indent=4))  # Debug: Imprime a payload
+
     session = requests.Session()
     response = session.post(INVOKE_URL, headers=headers, json=payload)
-    while response.status_code == 202:
+    if response.status_code == 202:
         request_id = response.headers.get("NVCF-REQID")
         fetch_url = FETCH_URL_FORMAT + request_id
         response = session.get(fetch_url, headers=headers)
-    response.raise_for_status()
+    response.raise_for_status()  # Isso lançará uma exceção se o status não for 200
     response_body = response.json()
     if response_body.get("choices"):
         assistant_message = response_body["choices"][0]["message"]["content"]
@@ -45,6 +45,7 @@ def call_nvidia_api(history, system_message, max_tokens, temperature, top_p):
     else:
         return "Desculpe, ocorreu um erro ao gerar a resposta."
 
+
 def chatbot_submit(message, chat_history, system_message, max_tokens_val, temperature_val, top_p_val):
     """Submits the user message to the chatbot and updates the chat history."""
     print("Updating chatbot...")
@@ -60,13 +61,13 @@ def chatbot_submit(message, chat_history, system_message, max_tokens_val, temper
 
     return assistant_message, chat_history
 
+system_msg = gr.Textbox(BASE_SYSTEM_MESSAGE, label="System Message", placeholder="System prompt.", lines=5)
+max_tokens = gr.Slider(20, 1024, label="Max Tokens", step=20, value=1024)
+temperature = gr.Slider(0.0, 1.0, label="Temperature", step=0.1, value=0.2)
+top_p = gr.Slider(0.0, 1.0, label="Top P", step=0.05, value=0.7)
 # Gradio interface setup
 with gr.Blocks() as demo:
     chat_history_state = gr.State([])
-    system_msg = gr.Textbox(BASE_SYSTEM_MESSAGE, label="System Message", placeholder="System prompt.", lines=5)
-    max_tokens = gr.Slider(20, 1024, label="Max Tokens", step=20, value=1024)
-    temperature = gr.Slider(0.0, 1.0, label="Temperature", step=0.1, value=0.2)
-    top_p = gr.Slider(0.0, 1.0, label="Top P", step=0.05, value=0.7)
     chatbot = gr.ChatInterface(
         fn=chatbot_submit,
         additional_inputs=[system_msg, max_tokens, temperature, top_p],