chatglm3-6B: benchmark results
Provider: Zhipu. Released 2023-10-25. Access: Open.
Unified ELO 1349 ± 1, rank #1270 of 1392 rated models, from 28 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| SALAD-Bench Attack | 12.48 | Safety Score (%) | 51.5 |
| T-Eval | 51.4 | Overall Score (%) | 50 |
| Open Chinese LLM - C-Eval Semantic | 68.8 | Accuracy (%) | 46.9 |
| SALAD-Bench | 51.47 | Average Safety Score (%) | 42.4 |
| SALAD-Bench Base | 90.45 | Safety Score (%) | 42.4 |
| Open Chinese LLM - TruthfulQA MC | 52.64 | Accuracy (%) | 39.2 |
| InfiBench | 28.23 | Score (%) | 27.6 |
| Open PL LLM - Multiple Choice | 0.18 | Average Multiple-Choice Score (%) | 21 |
| Open PL LLM Leaderboard | 2.76 | Average Score (%) | 17.8 |
| Open Chinese LLM - CMMLU | 46.11 | Accuracy (%) | 15.7 |
| Open PL LLM - Generative | 4.74 | Average Generative Score (%) | 15.4 |
| Open Chinese LLM Leaderboard | 47.86 | Average Score (%) | 12.8 |
Interactive version: theaggregate.ai/model?slug=chatglm3-6b · How It Works · Data refreshed daily, snapshot 2026-09-05.