GLM-4.6V (Non-reasoning): benchmark results
GLM-4.6V evaluated with reasoning disabled. Provider: Zhipu. Released 2025-12-09. Access: Open.
Unified ELO 1477 ± 1, rank #985 of 1761 rated models, from 39 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| UGI - Natural Intelligence | 32.31 | NatInt Score | 78.1 |
| UGI - Writing | 35.21 | Writing Score | 55.6 |
| UGI Leaderboard | 34.72 | UGI Score | 50.9 |
| AA Omniscience | -37.63 | Score | 49.7 |
| AA Omniscience - Software Engineering (SWE) | 23.2 | Accuracy (%) | 43.1 |
| AA Omniscience - Health | 18.6 | Accuracy (%) | 42.7 |
| AA TAU-2 Bench | 30.7 | Accuracy (%) | 37.8 |
| AA Omniscience - Science, Engineering & Mathematics | 24.8 | Accuracy (%) | 37.1 |
| AA-Omniscience Accuracy | 17.4 | Accuracy (%) | 36.8 |
| AA Omniscience - Humanities & Social Sciences | 16.8 | Accuracy (%) | 36 |
| AA Global-MMLU-Lite - Yoruba | 45.33 | Accuracy (%) | 35.1 |
| Artificial Analysis Intelligence Index | 5.24 | Intelligence Index | 35 |
Interactive version: theaggregate.ai/model?slug=glm-4-6v-non-reasoning · How It Works · Data refreshed daily, snapshot 2026-09-05.