vllm clarification

README.md

@@ -37,6 +37,8 @@ For more details, read the paper: [SafetyAnalyst: Interpretable, transparent, an
 
 ## How to Use HarmReporter
 
+Outputs from HarmReporter can be generated using the following code snippet:
+
 ```python
 from transformers import AutoTokenizer, AutoModelForCausalLM
 
@@ -49,6 +51,8 @@ input_tokenized = tokenizer.apply_chat_template(text_input, return_tensors="pt")
 output = model.generate(input_tokenized, max_new_tokens=19000)
 ```
 
+However, due to the extensive lengths of the harm trees generated by HarmReporter, we recommend using the [vllm](https://github.com/vllm-project/vllm) library to generate the outputs. 
+
 ## Intended Uses of HarmReporter
 
 - Harmfulness analysis: HarmReporter can be used to analyze the harmfulness of a given prompt in the hypothetical scenario that an AI language model provides a helpful answer to the prompt. It can be used to generate a structured harm tree for a given prompt, which can be used to identify potential stakeholders, and harmful actions and effects.