The first domestic company, 360 Intelligent Brain passed the trusted AIGC large language model function evaluation of China Academy of Information and Communications Technology

China-Singapore Jingwei, June 12th On June 12, 360 Intelligent Brain’s 360GPT-S2-V8 model product won the "Trusted AIGC Large Language Model" from the China Academy of Information and Communications Technology (hereinafter referred to as "CAICT") "Basic Ability" evaluation report (hereinafter referred to as the "report"), the report shows that all the required items have been passed. According to reports, 360 Intelligent Brain has passed the authoritative evaluation of the China Academy of Information and Communications Technology, becoming the first large-scale model product to receive certification in China.

China Academy of Information and Communications Technology is a scientific research institution directly under the Ministry of Industry and Information Technology. The Cloud Computing and Big Data Institute of China Academy of Information and Communications Technology (hereinafter referred to as "CAICT Cloud Institute") attaches great importance to the development trend of AIGC technology application and further promotes related work. In the process, the Cloud Institute of China Academy of Information and Communications Technology is currently relying on the "Content Technology Industry Promotion Array" and "Metaverse Innovation Exploration Array" to continuously promote the formulation of the "Trusted AIGC Series Standards". This evaluation report of 360 Intelligence is based on the "Trusted AIGC-Large Language Model Basic Capability (Function) Evaluation Method" in the "Trusted AIGC Series Standards".

The report shows that in addition to the 5 optional assessment contents, 360 Intelligent Brain has passed all 10 parts and 36 basic ability (function) assessments, and has the corresponding functions, including generation and creation, multi-round dialogue, Key areas related to large models such as coding ability, logic and reasoning, knowledge question and answer, reading comprehension, text classification, translation, text rewriting, multi-modal (text to image) capabilities, etc., can generate and create, multiple rounds of dialogue and accurately Understand functions such as conversational topics, fields, and emotions, and possess conversational reasoning and referential discrimination abilities, etc.

This is not the first time that 360 Intelligent Brain has appeared in third-party evaluations. In May, in an evaluation result of the Chinese General Large Model Benchmark (SuperCLUE), 360 Intelligent Brain ranked first among domestic large models in multiple capabilities. ; At the same time, the JioNLP large language model evaluation stated that 360 Intelligent Brain is "currently the best large language model in China."

It is reported that 360 Intelligent Brain is a large-scale self-developed cognitive intelligence general model, which has been gradually implemented and opened for internal testing in its 360 search, browser, office, intelligent hardware and other scenarios to further empower advantageous scenarios. According to public information, the application conference of the 360 ​​Intelligent Brain model is scheduled to be held in Beijing on June 13, 2023.

The above is the detailed content of The first domestic company, 360 Intelligent Brain passed the trusted AIGC large language model function evaluation of China Academy of Information and Communications Technology. For more information, please follow other related articles on the PHP Chinese website!