Qwen 1.5 32B Chat: benchmark results
Alibaba's dense 32B chat model in the Qwen1.5 line (32K context, April 2024), released under the custom commercial-use Qwen license. Provider: Alibaba. Released 2024-04-02. Access: Open.
Unified ELO 1486 ± 1, rank #751 of 1392 rated models, from 120 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Chinese LLM - C-Eval Semantic | 90.78 | Accuracy (%) | 97 |
| Open Chinese LLM - HellaSwag | 70.43 | Accuracy (%) | 94.7 |
| Open Chinese LLM - ARC Challenge | 60.67 | Accuracy (%) | 89.3 |
| Open Chinese LLM - TruthfulQA MC | 60.87 | Accuracy (%) | 85.2 |
| Open LLM Leaderboard - BBH | 44.55 | Score | 84.9 |
| Open Japanese LLM - Xlsum JA Bert Score JA F1 | 70.34 | Score (%) | 84.5 |
| BiGGen-Bench | 3.81 | Average Score (1-5) | 83.3 |
| Open LLM Leaderboard - MMLU-Pro | 38.41 | Score | 82.7 |
| Open Chinese LLM - WinoGrande | 67.4 | Accuracy (%) | 82.5 |
| Open Japanese LLM - Xlsum JA RougeLsum | 25.5 | Score (%) | 82.5 |
| Open Japanese LLM - Xlsum JA Rouge1 | 30.12 | Score (%) | 82.3 |
| Open Chinese LLM - CMMLU | 68 | Accuracy (%) | 78.6 |
Interactive version: theaggregate.ai/model?slug=qwen-1-5-32b-chat · How It Works · Data refreshed daily, snapshot 2026-09-05.