Today, Shangtang Technology released a new 5.0 model, using the MOE hybrid expert architecture, based on more than 10TB tokens training, 200K inference context windows, and fully benchmarking GPT-4 Turbo.
商汤发布日日新5.0大模型
On the day of the release of Shangtang, the new 5.0 big model
The translation is provided by third-party software.
The above content is for informational or educational purposes only and does not constitute any investment advice related to Futu. Although we strive to ensure the truthfulness, accuracy, and originality of all such content, we cannot guarantee it.