GLM-5.3 Flash (High): benchmark results
Provider: Zhipu. Released 2026-08-20. Access: Open.
Aggregate Rating 1765 ± 15, rank #210 of 2170 rated models, from 48 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| IndicBankBench - Confusing Intent | 71.7 | Strict pass^3 (%; 53 cases passed out of the 53 cases of the | 100 |
| IndicBankBench - Context Switching | 84.09 | Strict pass^3 (%; 44 cases passed out of the 44 cases of the | 100 |
| IndicBankBench - Multi-Tool Agentic | 39.29 | Strict pass^3 (%; 112 cases passed out of the 112 cases of t | 90 |
| IndicBankBench - Wrong Info | 38.03 | Strict pass^3 (%; 142 cases passed out of the 142 cases of t | 85 |
| Korean CSAT 2026 (Easy Mode) - Physics I | 42 | Points (out of 50) | 79 |
| IndicBankBench - Happy Path | 84.62 | Strict pass^3 (%; 39 cases passed out of the 39 cases of the | 75 |
| IndicBankBench - Third-Party Request Refusal | 92.31 | Strict pass^3 (%; 13 cases passed out of the 13 cases of the | 75 |
| DuelLab GameBench 2 - Reasoning Variants | 61.38 | GameBench2 normalized score (0–100) | 72.6 |
| LLM Chess (Saplin) | 571.68 | Native anchored-opponent Elo (signed points) | 69.4 |
| Context Arena | 63.43 | Average Score (%) | 68.9 |
| SelfBench - pingdotgg/t3code | 64.1 | Task success (%) on 39 tasks rebuilt from the repository's m | 68.8 |
| Korean CSAT 2026 (Easy Mode) - Society and Culture | 42 | Points (out of 50) | 68.7 |
Interactive version: theaggregate.ai/model?slug=glm-5-3-flash-high · How It Works · Data refreshed daily, snapshot 2026-11-05.