Actualiza la inicialización del modelo a InferenceClient y añade la importación de InferenceClient desde huggingface_hub

app.py

@@ -1,4 +1,5 @@
 from smolagents import CodeAgent, HfApiModel, load_tool, tool
+from huggingface_hub import InferenceClient
 import datetime
 import requests
 import pytz
@@ -79,12 +80,16 @@ web_searcher = DuckDuckGoSearchTool(max_results=5)
 # If the agent does not answer, the model is overloaded, please use another model or the following Hugging Face Endpoint that also contains qwen2.5 coder:
 # model_id='https://pflgm2locj2t89co.us-east-1.aws.endpoints.huggingface.cloud' 
 
-model = HfApiModel(
-    max_tokens=2096,
-    temperature=0.5,
-    model_id='Qwen/Qwen2.5-Coder-32B-Instruct',
-    custom_role_conversions=None,
-    token=HF_TOKEN
+# model = HfApiModel(
+#     max_tokens=2096,
+#     temperature=0.5,
+#     model_id='Qwen/Qwen2.5-Coder-32B-Instruct',
+#     custom_role_conversions=None,
+#     token=HF_TOKEN
+# )
+model = InferenceClient(
+    model="Qwen/Qwen2.5-Coder-32B-Instruct",
+    api_key=HF_TOKEN
 )