Update README.md

README.md

@@ -218,6 +218,8 @@ for output in outputs:
     print(f"Decoder prompt: {prompt!r}, Generated text: {generated_text!r}")
 ```
 
+*Note:* we recommend using the default prompt that extracts bounding boxes, classes, and text in markdown formatting for all use cases (`</s><s><predict_bbox><predict_classes><output_markdown>`). If necessary, optionally the prompt that omits text extraction and only outputs bounding boxes and classes could be used: `</s><s><predict_bbox><predict_classes><output_no_text>`.
+
 Nemotron-Parse-v1.1 is also available as an [optimized NIM container](https://build.nvidia.com/nvidia/nemotron-parse).
 
 ## Nemotron-Parse-v1.1 Output Examples