Qwen 3 1.7B (Thinking) — benchmark results
Provider: Alibaba. Released 2025-04-28. Access: Open.
Unified ELO 1392 ± 35, rank #1263 of 1776 rated models, from 41 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA MATH-500 | 89.4 | Accuracy (%) | 64.9 |
| AA AIME 2025 | 38.67 | Accuracy (%) | 40 |
| AA LiveCodeBench | 30.79 | Pass@1 (%) | 36.5 |
| BRIDGE Medical Leaderboard - CoT | 27.71 | Average Performance (%) | 33 |
| CritPt | 0 | Accuracy (self-reported) | 30.3 |
| AA TAU-2 Bench | 26.02 | Accuracy (%) | 30.1 |
| AA Humanity's Last Exam | 4.77 | Accuracy (%) | 28.3 |
| AA CritPt | 0 | Accuracy (%) | 27.1 |
| BRIDGE Medical Leaderboard | 28.95 | Average Performance (%) | 23.6 |
| BRIDGE Medical Leaderboard - Zero-Shot | 26.28 | Average Performance (%) | 23.6 |
| BRIDGE Medical Leaderboard - Few-Shot | 32.87 | Average Performance (%) | 20.8 |
| AA MMLU-Pro | 56.97 | Accuracy (%) | 18.9 |
Interactive version: theaggregate.ai/model?slug=qwen-3-1-7b-thinking · How the rankings work · Data refreshed daily, snapshot 2026-07-22.