Update app.py

app.py

@@ -4,11 +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
 """
-#client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
-#client = InferenceClient("meta-llama/Llama-3.2-1B-Instruct")
-#client = InferenceClient("microsoft/Phi-3.5-mini-instruct")
-#client = InferenceClient("unsloth/Llama-3.2-1B-Instruct")
-client = InferenceClient("mlx-community/Hermes-3-Llama-3.1-70B-8bit")
+client = InferenceClient("mlx-community/Hermes-3-Llama-3.1-70B-8bit")  # Garantir que o modelo seja compatível com `text-generation`
 
 
 def respond(
@@ -29,18 +25,18 @@ def respond(
 
     messages.append({"role": "user", "content": message})
 
-    response = ""
-
-
-    mensagens = client.chat_completion(
-        messages,
+    # Alterar a requisição para utilizar um modelo de `text-generation`
+    response = client.text_generation(
+        inputs=messages[-1]['content'],
         max_tokens=max_tokens,
         temperature=temperature,
-        top_p=top_p,
+        top_p=top_p
     )
-    response = mensagens.choices[0].message.content
 
-    return response
+    # Assumindo que o modelo de texto retorna a resposta como uma string diretamente
+    response_text = response['generated_text']  # Adapte de acordo com a estrutura da resposta
+
+    return response_text
 
 
 """
@@ -64,4 +60,4 @@ demo = gr.ChatInterface(
 
 
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch()