In addition to open-source models, Skywork-13B series large models will also open-source high-quality Chinese corpus data set Skypile Chinese-web-text-150B for 600GB and 150B Tokens, which is currently one of the largest open-source Chinese data sets.