GLM-5.2 benchmark ranking on every AI leaderboard

open weights · top 10 on 0 of 7 · released Jun 13, 2026

Z.ai (Zhipu)Correct

Rank	Board	Category	Score	Of	Over time
15	OpenRouterUsage · as “z-ai/glm-5.2-20260616”	Usage	1.7T tok	19
18	WebDev ArenaCoding · as “glm-5.2-max”	Coding	1,600	100
19	GPQA DiamondReasoning · as “glm-5.2_max”	Reasoning	91.9%	100
32	Text ArenaChat · as “glm-5.2-max”	Chat	1,472	100
37	ARC-AGI-2Reasoning · as “GLM-5.2”	Reasoning	22.8%	78
41	LiveBenchReasoning · as “glm-5.2”	Reasoning	73.2%	59
43	Epoch ECIReasoning · as “GLM-5.2”	Reasoning	151.8	100

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