Introducing BrainTrace's lightweight flagship model, benchmarked against DeepSeek, supported by 100,000 domestic cards

Perceive Beating AI News Flash: Zhìpǔ has released a lightweight flagship model GLM-5.3 Flash aimed to benchmark against DeepSeek V4 Flash, utilizing over 100,000 domestic chip cluster for the model's inference service.

It is understood that its computing power supplier may be from Huawei, Moore Threads, and SeaLight, but Zhìpǔ officials have not commented on this information. Zhìpǔ stated in technical documents that its cluster's "hardware efficiency and unit token cost have reached a level equivalent to mainstream NVIDIA GPUs," but did not specify the exact model. (Stay tuned)