Llama3-ChatQA-1.5-70B is an advanced conversational question-answering and retrieval-augmented generation (RAG) model developed by NVIDIA. Based on the Llama-3 base model, it has employed enhanced training methods and especially enhanced its capabilities in table and arithmetic calculations. It comes in two variants: Llama3-ChatQA-1.5-8B and Llama3-ChatQA-1.5-70B. The model has achieved outstanding results in multiple conversational question-answering benchmarks, demonstrating its efficient ability to handle complex conversations and generate relevant responses.