摩尔线程MTT S5000完成GLM-5全流程适配

Group 1 - The core announcement is that Moole Technology (688795.SH) has completed Day-0 adaptation and verification of the GLM-5 model from Zhizhu AI (02513.HK) on its flagship AI training and inference GPU, the MTT S5000 [1] - The GLM-5 model, released on February 11, is positioned as a top-tier coding model with a 20% performance improvement over its predecessor, featuring advanced capabilities in Agentic Engineering for end-to-end development [1][2] - GLM-5 ranks fourth globally and first among open-source models in the authoritative Artificial Analysis rankings, achieving state-of-the-art scores in key benchmarks [2] Group 2 - The MTT S5000 GPU is designed specifically for large model training, inference, and high-performance computing, with a maximum AI computing power of 1000 TFLOPS and 80GB of memory [2] - The memory bandwidth of the MTT S5000 reaches 1.6TB/s, and inter-card bandwidth is 784GB/s, supporting full precision calculations from FP8 to FP64 [2]