Qwen 3 0.6B (Thinking) — benchmark results
Qwen 3 0.6B evaluated with thinking enabled. Provider: Alibaba. Released 2025-04-28. Access: Open.
Unified ELO 1307 ± 26, rank #1517 of 1776 rated models, from 46 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA Humanity's Last Exam | 5.7 | Accuracy (%) | 41.3 |
| AA MATH-500 | 75 | Accuracy (%) | 38.1 |
| CritPt | 0 | Accuracy (self-reported) | 30.3 |
| AA CritPt | 0 | Accuracy (%) | 27.1 |
| UGI - Willingness (W/10) | 3 | W/10 Score | 22.2 |
| AA TAU-2 Bench | 21.05 | Accuracy (%) | 20.8 |
| AA AIME 2025 | 18 | Accuracy (%) | 20.6 |
| BRIDGE Medical Leaderboard - CoT | 18.95 | Average Performance (%) | 13.2 |
| AA Omniscience - Software Engineering (SWE) - R | 2 | Accuracy (%) | 11.9 |
| AA Omniscience - Software Engineering (SWE) - PHP | 10 | Accuracy (%) | 11.5 |
| BRIDGE Medical Leaderboard | 21.4 | Average Performance (%) | 11.3 |
| BRIDGE Medical Leaderboard - Zero-Shot | 20.38 | Average Performance (%) | 11.3 |
Interactive version: theaggregate.ai/model?slug=qwen-3-0-6b-thinking · How the rankings work · Data refreshed daily, snapshot 2026-07-22.