April 15 news, Zhipu AI released a new generation of open source model GLM-4-32B-0414 series, including pedestal, inference, contemplation model weights, and comply with the MIT License. All models in the current series can be accessed through "Z.ANi". According to reports, the inference model GLM-Z1-Air/AirX-0414 model inference speed can be up to 200 Tokens/ second (measured on MaaS platform bigmodel.cn), which is also the fastest business model in China so far. And the price of the GLM-Z1-Air-0414 is only 1/30 of that of the DeepSeek-R1
Share this post
Zhipu AI released a new generation of open source model, the maximum speed version of 200tokens/ second
Share this post
April 15 news, Zhipu AI released a new generation of open source model GLM-4-32B-0414 series, including pedestal, inference, contemplation model weights, and comply with the MIT License. All models in the current series can be accessed through "Z.ANi". According to reports, the inference model GLM-Z1-Air/AirX-0414 model inference speed can be up to 200 Tokens/ second (measured on MaaS platform bigmodel.cn), which is also the fastest business model in China so far. And the price of the GLM-Z1-Air-0414 is only 1/30 of that of the DeepSeek-R1
.