Z.aiopen weightsCurrentReasoning
GLM-5.3
Weights released Aug 28, 2026.
Aug 14, 2026
$1.4 / $4.4
$2.15
1M
Benchmark scores
| Benchmark | Score | Δ vs prior | Rank (current) | Configuration | Source |
|---|---|---|---|---|---|
Terminal-Bench 2.1 | 71.5% | #12 of 22 | leaderboard | benchlm.ai ↗ | |
GPQA Diamond | 88.1% | #14 of 21 | Vals AI run, max reasoning · independent run | vals.ai ↗ | |
Humanity's Last Exam | 42.3% | #17 of 31 | Artificial Analysis run, no tools (2,158 text-only questions), max effort; Z.ai reports 62.5 with tools · independent run | artificialanalysis.ai ↗ | |
MMLU-Pro | 86.8% | #6 of 12 | Vals AI run, max reasoning · independent run | vals.ai ↗ | |
LMArena Text | 1475 | #4 of 4 | max · leaderboard | benchlm.ai ↗ |