Recently, Moore Thread and Shi Yin Intelligence announced a strategic partnership to jointly promote the application of domestic full-function GPUs in industry-wide large-model solutions. The two parties have completed the adaptation work of Moore Thread's Kua E Zhi Suan cluster and Shi Yin Intelligence's "One Leaf Light Boat" multi-modal large model, integrating their respective advantages in the field of artificial intelligence to provide a more intelligent and efficient service experience.

Shi Yin Intelligence's "One Leaf Light Boat" multi-modal large model boasts industry-leading multi-modal and multi-task general artificial intelligence capabilities, covering key technical fields such as natural language processing, intelligent content generation, voice, and visual recognition. Through quantitative compression technology, the model has constructed multiple commercial-grade models ranging from 0.7B to 45B, meeting the diverse business needs in practical application scenarios. The Mixture of Experts (MoE) model based on the recurrent gating network demonstrates high configurability and universality, providing an efficient platform for AI application practices.

With the powerful computing power support of Moore Thread's Kua E Zhi Suan cluster, Shi Yin Intelligence has completed the training and testing of a 7 billion parameter large model. The test results indicate a high degree of compatibility and adaptation, with training efficiency meeting expectations, fully validating the high performance of the intelligent computing cluster. This deep integration will further enhance the application scenarios and service capabilities of the large model, providing customers with more powerful and customized solutions.

Shi Yin Intelligence's products and services have been widely applied in various scenarios such as government services, financial securities, smart offices, and intelligent manufacturing, providing strong support for the intelligent transformation and upgrading of industries. In the future, Moore Thread and Shi Yin Intelligence will continue to deepen their cooperation, promoting the integration of domestic computing power and domestic large models, and jointly creating in-depth industry solutions to better empower various industries.