Zhipu's GLM-5.3-Highspeed reportedly reaches 490 tokens/s in Zcode

Zhipu's GLM-5.3-Highspeed model is reportedly available to random Zcode 3.15.1 subscribers, with one user reporting a speed of 490 tokens/s.

That is well above the 200 tokens/s top speed of GLM-5.3-FlashX released last month, and Artificial Analysis measured GLM-5.3-Flash at about 60 tokens/s on the Z.ai API. QbitAI could not test the model because it was not yet available to its account.