Skip to content
Leaders Board.
Models/Z.ai (Zhipu)Open weights

GLM-5.2

Ranked on the boards below.
-
Overall rank
-
Podium score
0
Board wins
Jun 13, 2026
Released
·Chat·Reasoning·Coding·Usage

Chat

Which model people prefer in blind head-to-head conversations.

32
Text Arena
1,472 · of 100
as “glm-5.2-max”

Reasoning

Hard questions with checkable answers: science, math, puzzles, expert exams.

19
GPQA Diamond
91.9% · of 100
as “glm-5.2_max”
37
ARC-AGI-2
22.8% · of 78
as “GLM-5.2”
41
LiveBench
73.2% · of 59
as “glm-5.2”
43
Epoch ECI
151.8 · of 100
as “GLM-5.2”

Coding

Fixing real repositories, building web apps, editing code.

18
WebDev Arena
1,600 · of 100
as “glm-5.2-max”

Usage

What developers actually send tokens to.

15
OpenRouter
1.7T tok · of 19
as “z-ai/glm-5.2-20260616”

More from Z.ai (Zhipu)

GLM-5.3GLM 4.6GLM 5.3 FlashGLM-5.1GLM-5GLM-4.7GLM 4.5GLM-5v Turbo

Scores as published by each board on the capture date. Model names and logos belong to their owners; logos via logo.dev.

Weekly: the AI leaderboards with a new #1, Saturday mornings.