The requirement is to build a web application that evaluates the responses of different LLMmodels in the Question and Answering task. The models to compare are as follows$1.$ gpt$-3\mathrm{.}5-$turbo$2.$ gpt$-43.$ Llama$-2-70$b$-$chat$4.$ Falcon$-40$b$-$instruct