Qwen3 30B A3B Thinking 8-bit: benchmark results
Provider: Alibaba.
Aggregate Rating 1622 ± 27, rank #817 of 10916 rated models, from 28 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Medmarks - M-ARC | 60.33 | Score (%) | 80.1 |
| Medmarks - Med-HALT Reasoning NOTA | 67.24 | Score (%) | 79.5 |
| Medmarks - PubMedQA | 76.8 | Score (%) | 72.7 |
| Medmarks - MedHallu Medium | 63.21 | Score (%) | 71.6 |
| Medmarks - MedCalc-Bench | 62.18 | Score (%) | 67 |
| Medmarks - LongHealth Task 1 | 88.58 | Score (%) | 66.5 |
| Medmarks - MMLU-Pro Health | 76.37 | Score (%) | 65.9 |
| Medmarks - Med-HALT Reasoning FCT | 78.63 | Score (%) | 65.9 |
| Medmarks - SuperGPQA Medicine Easy | 58.01 | Score (%) | 65.9 |
| Medmarks - MedQA | 88.56 | Score (%) | 64.2 |
| Medmarks - SuperGPQA Medicine Hard | 43.16 | Score (%) | 63.6 |
| Medmarks - Verifiable | 53.95 | Weighted Mean Win Rate (%) | 62.5 |
Interactive version: theaggregate.ai/model?slug=qwen3-30b-a3b-thinking-8-bit · How It Works · Data refreshed daily, snapshot 2026-11-08.