GLM-5.3 Benchmarks: Artificial Analysis Scores, Full Comparison Table, and Pricing
GLM-5.3 benchmarks today: 45 on Artificial Analysis Intelligence Index v4.3.2, 28.3 on Terminal-Bench 3.0, 66.9 on DeepSWE v1.1, and 1769 Elo on GDPval-AA v2 — the highest of any open-weight model. The famous “60” from August is the same model on an older index version. That is the entire controversy compressed into two sentences. Everything below is the evidence behind it: the vendor table Z.ai published, the independent Artificial Analysis run, the price you actually pay per unit of work, and the capability gaps that did not close. ...