GLM-5.3-FlashX is the high-speed variant of Z.ai's GLM-5.3-Flash, a native multimodal model delivering inference speeds of up to 200 tokens/s. Built on the same hybrid sparse and linear attention architecture...
Rank
#68
Down 13% week over week
Tokens
199.1B
2026-10-15
Requests
3.5M
3,465,278
Tool calls
-
Input / M
$0.37
Output / M
$1.25
Cache / M
$0.09
Context
1.0M
1,048,576 tokens