GLM-5.3 from Z.ai scores 60 on the Artificial Analysis Intelligence Index, tying Kimi K3 for the top spot among open models. The model shows major gains in agentic tasks, reaching an Elo score of 1,770 on the GDPval-AA v2 benchmark, second only to Claude Opus 5. Priced at $0.68 per task, GLM-5.3 is 19 percent cheaper than Kimi K3. Z.ai is delaying open weight release by two weeks to strengthen security controls.
No score is assigned. Sources and their independence are shown in the citation chain below.