fix model task

app.py

@@ -37,6 +37,7 @@ translation_pipeline = pipeline(
 def get_llm_hf_inference(model_id=model_id, max_new_tokens=128, temperature=temperature):
     llm = HuggingFaceEndpoint(
         repo_id=model_id,
+        task="text-generation",
         max_new_tokens=max_new_tokens,
         temperature=temperature,
         token = os.getenv("HF_TOKEN")