GLM-5.3-Prime is the high-speed variant of Z.ai's GLM-5.3, inheriting its full capabilities while delivering 1.5–2× the output throughput through inference acceleration. It supports text input and output with a 1M-token...
Rank
#94
Up 25% week over week
Tokens
76.4B
2026-10-15
Requests
988.5K
988,520
Tool calls
-
Input / M
$2.80
Output / M
$8.80
Cache / M
$0.56
Context
1M
1,000,000 tokens