granite didn't work as intended .. switch back to llama

app.py

@@ -4,8 +4,8 @@ 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("ibm-granite/granite-3.1-2b-instruct") # specifically for long contexts.
-# client = InferenceClient("meta-llama/Llama-3.2-3B-Instruct") # too small context window
+# client = InferenceClient("ibm-granite/granite-3.1-2b-instruct") # specifically for long contexts.
+client = InferenceClient("meta-llama/Llama-3.2-3B-Instruct") # too small context window
 # client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
 
 _sys_msg = """