On GLM-5.3, the endpoints with the fastest streaming take 18 to 26 seconds to return the first token, while most others start in 1 to 3 seconds. Once that wait counts toward the full request time, they fall down the speed rankings.
Artificial Analysis says fastest GLM-5.3 endpoints lag on total response time
AIArtificial Analysis reports that the GLM-5.3 endpoints with the fastest streaming take 18 to 26 seconds to return the first token, while most others take 1 to 3 seconds. Once that wait is counted toward total request time, these endpoints drop in the speed rankings.
Source details
Artificial Analysis · x.comPublished
Original title: On GLM-5.3, the endpoints with the fastest streaming take 18 to 26 seconds to return the first token, while most others start in 1 to 3 s...