Qwen3 30B A3B Thinking 4-bit: benchmark results
Provider: Alibaba.
Aggregate Rating 1609 ± 27, rank #934 of 10916 rated models, from 28 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Medmarks - Med-HALT Reasoning NOTA | 68.45 | Score (%) | 83 |
| Medmarks - PubMedQA | 77 | Score (%) | 76.1 |
| Medmarks - SuperGPQA Medicine Hard | 44.55 | Score (%) | 68.2 |
| Medmarks - M-ARC | 49 | Score (%) | 65.9 |
| Medmarks - MedCalc-Bench | 59.73 | Score (%) | 62.5 |
| Medmarks - Med-HALT Reasoning FCT | 76.37 | Score (%) | 61.4 |
| Medmarks - MedQA | 87.67 | Score (%) | 61.4 |
| Medmarks - MedHallu Medium | 60.85 | Score (%) | 59.1 |
| Medmarks - SuperGPQA Medicine Easy | 55.92 | Score (%) | 58 |
| Medmarks - CareQA EN | 88.72 | Score (%) | 56.8 |
| Medmarks - HEAD-QA v2 | 86.44 | Score (%) | 56.8 |
| Medmarks - MedHallu Easy | 66.79 | Score (%) | 56.8 |
Interactive version: theaggregate.ai/model?slug=qwen3-30b-a3b-thinking-4-bit · How It Works · Data refreshed daily, snapshot 2026-11-08.