GLM-5.3 Flash (Low): benchmark results
Provider: Zhipu. Released 2026-08-20. Access: Open.
Aggregate Rating 1687 ± 24, rank #426 of 2286 rated models, from 33 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| K-Bench | 98.36 | Overall (%, Default v1 prompt) | 81.1 |
| CMDB-1500 (Image) | 90.67 | Accuracy (%), fixed denominator of 300 questions (invalid an | 77.8 |
| K-Bench (Therapeutic v0) | 97.96 | Overall (%, Therapeutic v0 prompt) | 75.8 |
| K-Bench - Risk (Therapeutic v0) | 91.83 | Risk Score (%, Therapeutic v0 prompt) | 62.1 |
| StudyArena Text - Weighted score | 1225 | Weighted score | 61.5 |
| Context Arena | 59.1 | Average Score (%) | 61 |
| K-Bench - Risk | 91.69 | Risk Score (%, Default v1 prompt) | 52.5 |
| ARC-AGI-2 | 27.92 | Accuracy (%) | 48.2 |
| DuelLab GameBench 2 - Reasoning Variants | 46.83 | GameBench2 normalized score (0–100) | 44.8 |
| CMDB-1500 (Text) | 75.75 | Accuracy (%), fixed denominator of 1,200 questions (invalid | 44.4 |
| CAE-Bench RC7 - Fixed Problems | 73.55 | Score (0-100), 24 tasks | 37.5 |
| ARC-AGI-1 | 47 | Accuracy (%) | 30.9 |
Interactive version: theaggregate.ai/model?slug=glm-5-3-flash-low · How It Works · Data refreshed daily, snapshot 2026-11-20.