Qwen 3 30B A3B (Thinking): benchmark results
Alibaba Qwen 3 30B A3B evaluated with thinking enabled. Provider: Alibaba. Released 2025-04-29. Access: Open.
Aggregate Rating 1551 ± 9, rank #875 of 2170 rated models, from 189 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| SuperCLUE General (May 2025) - Science Reasoning | 50.5 | Score | 87.2 |
| BRIDGE Medical Leaderboard - CoT | 39.35 | Average Performance (%) | 86.5 |
| SuperCLUE General (May 2025) - Math Reasoning | 61.29 | Score | 84.6 |
| BRIDGE Medical Leaderboard - Zero-Shot | 40.93 | Average Performance (%) | 82.9 |
| Medmarks - M-ARC | 61.33 | Score (%) | 81.8 |
| Medmarks - PubMedQA | 77.47 | Score (%) | 80.7 |
| CodeMixQA | 20.84 | SimpleQA F1 (%), mean of 4 code-switching perturbations | 80 |
| CodeMixQA - Grammar Forcing (Src) | 21.45 | SimpleQA F1 (%) | 80 |
| CodeMixQA - Grammar Forcing (Tgt) | 20.44 | SimpleQA F1 (%) | 80 |
| CodeMixQA - Random Switching | 20.89 | SimpleQA F1 (%) | 80 |
| CodeMixQA - Selective Switching | 20.57 | SimpleQA F1 (%) | 80 |
| MADE | 45 | Macro-F1 (%, the 0-1 score times 100) over all 1,154 labels | 78.6 |
Interactive version: theaggregate.ai/model?slug=qwen-3-30b-a3b-thinking · How It Works · Data refreshed daily, snapshot 2026-11-05.