GLM-5.3 Flash (Low): benchmark results
Provider: Zhipu. Released 2026-08-20. Access: Open.
Unified ELO 1721 ± 29, rank #310 of 2066 rated models, from 16 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| K-Bench | 98.36 | Overall (%, Default v1 prompt) | 81.1 |
| 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 |
| Context Arena | 59.1 | Average Score (%) | 61 |
| ARC-AGI-2 | 27.92 | Accuracy (%) | 52.7 |
| K-Bench - Risk | 91.69 | Risk Score (%, Default v1 prompt) | 52.5 |
| DuelLab Overall | 35.1 | DuelLab Score | 50 |
| ARC-AGI-1 | 47 | Accuracy (%) | 35.3 |
| Bug Hunt Bench - VS Code Extension | 7 | Planted Bugs Fixed (out of 45) | 33.3 |
| NonoBench | 23.3 | Overall Accuracy (%) | 18.6 |
| Bug Hunt Bench | 9 | Planted Bugs Fixed (out of 105) | 10.6 |
| CursorBench 4.0 | 26.9 | Score (%) | 10 |
Interactive version: theaggregate.ai/model?slug=glm-5-3-flash-low · How It Works · Data refreshed daily, snapshot 2026-10-05.