Head to head

GLM-5.2 vs GLM 5.1

On a blended workload of one million input tokens and three million output tokens, GLM 5.1 is line-ball with GLM-5.2.

GLM-5.2GLM 5.1
Provider Together AI Fireworks AI
Input / MTok $1.4 $1.4
Output / MTok $4.4 $4.4
Cache read / MTok $0.26
Context window 262,144
Blended 1M in + 3M out $14.60 $14.60
Status ga ga

GLM-5.2 detail   GLM 5.1 detail