Qwen3-32B-AWQ: benchmark results
Provider: Alibaba. Access: Open.
Aggregate Rating 1627 ± 40, rank #777 of 10916 rated models, from 19 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| ReasonScape M12X | 626.1 | ReasonScore (hard) | 92.4 |
| Grip on LLMs - Dutch tinyARC | 81 | Accuracy (%; GP-IRT estimate from 100 tinyBenchmarks items, | 88.3 |
| Grip on LLMs - HonestCityBench - Incorrect Premise | 65 | Appropriate acknowledgement rate (%; LLM judge) | 88.3 |
| Avalon-ToM-Bench - Intention Attribution | 89.83 | Accuracy (%) | 84.6 |
| Grip on LLMs - HonestCityBench - Insufficient Information | 44 | Appropriate acknowledgement rate (%; LLM judge) | 80 |
| Avalon-ToM-Bench - Strategic Signaling | 77.94 | Accuracy (%) | 76.9 |
| Grip on LLMs - Dutch Factuality Average | 69 | Accuracy (%; mean of three GP-IRT estimates, Dutch machine t | 76.7 |
| Grip on LLMs - Dutch tinyTruthfulQA | 68 | Accuracy (%; GP-IRT estimate from 100 tinyBenchmarks items, | 76.7 |
| Grip on LLMs - Dutch tinyMMLU | 59 | Accuracy (%; GP-IRT estimate from 100 tinyBenchmarks items, | 73.3 |
| Avalon-ToM-Bench - Average | 78.49 | Accuracy (%; macro-average of four abilities, balanced true/ | 73.1 |
| Avalon-ToM-Bench - Tacit Coordination | 82.01 | Accuracy (%) | 69.2 |
| Grip on LLMs - HonestCityBench - Average | 30 | Appropriate acknowledgement rate (%; mean of five categories | 66.7 |
Interactive version: theaggregate.ai/model?slug=qwen3-32b-awq · How It Works · Data refreshed daily, snapshot 2026-11-08.