tensorrt_llm==0.18.0
datasets==2.14.6
evaluate~=0.4.1
rouge_score~=0.1.2
sentencepiece>=0.1.99