iFLYTEK released the iFLYTEK Spark Cognitive Model V1.5, which has undergone three major upgrades in knowledge question and answer, logical reasoning, multi-round dialogue, text generation, language understanding, numerical ability and code ability, and its comprehensive ability has been continuously improved.
According to the official introduction, the model has made a breakthrough in open-ended knowledge Q&A, which can provide users with more comprehensive and professional information. In addition, the logical reasoning and mathematical capabilities of iFLYTEK Xinghuo Cognitive Big Model V1.5 have been upgraded, which can handle complex scenarios such as farmers, wolves, sheep, and vegetables crossing the river, straight line intersection problems, and probability problems.
V1.5 also improves the ability to multi-round dialogue, which can complete specific and complex tasks, and brings new opportunities for human-machine co-creation. In addition, text generation ability increased by 7%, language comprehension ability increased by 8%, knowledge question answering ability increased by 24%, logical reasoning ability increased by 10%, numerical ability increased by 9%, and code ability increased by 8%.
iFLYTEK Xinghuo cognitive large model will continue to be upgraded this year, and it is planned to launch new features on June 9, August 15 and October 24, respectively, and reach the level of a general model comparable to ChatGPT by the end of the year.
Liu Qingfeng, Chairman of iFLYTEK, said that the model has been applied in four major fields: education, office and digital employees, bringing new opportunities for industrial transformation.
(8210289)