Huawei Ascend 950DT Hits 86% of Nvidia H200 Inference Performance, 1,951 Tokens/s Single-Card Throughput

According to ChainCatcher citing an internal China Mobile test report from October, Huawei's Ascend 950DT achieved 86% of Nvidia H200's inference performance on the GLM 5.2 benchmark. Testing conducted on October 10 using 64-card Ascend 950DT clusters showed single-card throughput of 1,951 tokens/s. The report estimated H200 throughput at approximately 2,200 tokens/s in the same scenario. Additionally, the material indicated Ascend 950DT's inference cost-effectiveness was approximately 30% higher than H200.
Disclaimer: This content is for informational purposes only and does not constitute investment advice or encourage the buying or selling of crypto assets. It is intended to provide information about industry developments and is not directly related to the services offered by Gate TR. Although information obtained from third-party sources has been carefully reviewed, Gate TR assumes no responsibility for any errors or omissions. Crypto assets involve high risks and may experience significant price fluctuations. Before making any investment decisions, you should assess your financial situation and risk tolerance, conduct your own research, and make decisions independently.