Kunlun Tech releases open source 13B high-quality commercial large model, ahead of Llama2 and Baichuan2
On October 30, Kunlun Tech announced the open source of the 10 billion level large language model Skywork-13B series, and a rare support for the open source 600GB, 150B Tokens of the super-high quality Chinese data set.
The Skywork-13B series currently includes two major models with 13 billion parameters: The skywork-13B-Base model and, the Skywork-13B-Math model, they have shown the best effect of the same scale model in CEVAL, GSM8K, and other authoritative evaluation and benchmark tests, and their Chinese ability is particularly excellent, and their performance in Chinese technology, finance, government and other fields is higher than other open source models.
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.
At the same time, the Kunlun Tech Skywork-13B series large model will be fully open for commercial use - developers do not need to apply for commercial use.
13 billion parameters, two large models, and one of the largest Chinese data sets, fully open for commercial use. Kunlun Tech Skywork-13B series large model is the industry's most thorough open source 10 billion high-quality commercial model.
The open-source of Skywork-13B series large models will provide the best technical support for the scenario application of large models and the development of open-source communities, reduce the commercial threshold of large models, promote the implementation of artificial intelligence technology in thousands of industries, contribute to the construction of artificial intelligence ecology, and join hands with the open source community to explore the unknown world and create a better future.