On December 22, the results of China's first official "Large Model Standard Compliance Evaluation" were announced. Alibaba Cloud Tongyi Qianwen has become one of the first four domestic large models to pass the evaluation, and has met the requirements of relevant national standards in terms of versatility and intelligence.
The "Large Model Standard Compliance Evaluation" was initiated by the China Electronics Standardization Institute, aiming to establish a list of China's large model standards and lead the healthy and orderly development of the artificial intelligence industry. The evaluation solicited the opinions of dozens of leading units in academia and industry, covering 38 specific evaluation dimensions to evaluate the versatility and intelligence of the large language model, and is an authoritative evaluation based on the official large model test benchmark.
Among the first batch of large models that passed the evaluation, Tongyi Qianwen is the only open-source model, which has a wide range of developer users and enterprise customers around the world, and its performance and security have been publicly tested on a large scale. After the open source on December 1, Tongyi Qianwen 72B achieved the best results of the open source model in 10 authoritative benchmark evaluations, and beat LLAMA2 to the top of the most authoritative HuggingFace list overseas, and then topped the OpenCompass list of the Shanghai Artificial Intelligence Laboratory in China, becoming the most powerful open source model recognized by the industry.
At present, Tongyi Qianwen APP can be experienced in the major app stores of Apple and Android, providing dozens of practical functions such as text dialogue, voice dialogue, literary analysis, foreign language and classical Chinese translation, PPT outline assistant, Xiaohongshu copywriting, etc.
Shiri. Proofread by Li Haihui.