Update app.py

app.py

@@ -4,8 +4,7 @@ from huggingface_hub import InferenceClient
 """
 For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
 """
-# Criando o cliente para interagir com o modelo no Hugging Face.
-client = InferenceClient("qwen2.5:0.5b")
+client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
 
 
 def respond(
@@ -16,7 +15,6 @@ def respond(
     temperature,
     top_p,
 ):
-    # Prepara as mensagens para a API
     messages = [{"role": "system", "content": system_message}]
 
     for val in history:
@@ -27,46 +25,24 @@ def respond(
 
     messages.append({"role": "user", "content": message})
 
-    response = ""  # Resposta acumulada
+    response = ""
 
-    try:
-        # Chama a API de completamento com streaming
-        # A API do Hugging Face usa o método `client.chat_completion`.
-        response_stream = client.chat_completion(
-            messages=messages,
-            max_tokens=max_tokens,
-            temperature=temperature,
-            top_p=top_p,
-            stream=True
-        )
+    for message in client.chat_completion(
+        messages,
+        max_tokens=max_tokens,
+        stream=True,
+        temperature=temperature,
+        top_p=top_p,
+    ):
+        token = message.choices[0].delta.content
 
-        for message in response_stream:
-            # Verifica se a resposta contém o conteúdo esperado
-            if 'choices' not in message or len(message['choices']) == 0 or 'delta' not in message['choices'][0]:
-                raise ValueError("Resposta inesperada do modelo.")
-
-            token = message['choices'][0]['delta']['content']
-            response += token  # Acumula o conteúdo
-
-            # Retorna a resposta incrementalmente
-            yield response
-
-    except ValueError as e:
-        print(f"Erro de valor: {e}")
-    except ConnectionError as e:
-        print(f"Erro de conexão: {e}")
-    except TimeoutError as e:
-        print(f"Erro de tempo: {e}")
-    except Exception as e:
-        print(f"Erro inesperado: {e}")
-
-    return response  # Retorna a resposta final ao final do processamento
+        response += token
+        yield response
 
 
 """
 For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
 """
-# Criando a interface Gradio
 demo = gr.ChatInterface(
     respond,
     additional_inputs=[
@@ -85,4 +61,4 @@ demo = gr.ChatInterface(
 
 
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch()