Qwen 3 1.7B (Thinking): benchmark results
Provider: Alibaba. Released 2025-04-29. Access: Open.
Aggregate Rating 1398 ± 12, rank #1467 of 2168 rated models, from 135 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Token Rewards Evaluation - AIME24 | 46.7 | Pass@1 (%) | 85.7 |
| Token Rewards Evaluation - AIME25 | 35.4 | Pass@1 (%) | 85.7 |
| Token Rewards Evaluation - AMC23 | 85.3 | Pass@1 (%) | 85.7 |
| Token Rewards Evaluation - MATH-500 | 91.9 | Pass@1 (%) | 85.7 |
| Token Rewards Evaluation - MinervaMath | 49.2 | Pass@1 (%) | 85.7 |
| Token Rewards Evaluation - OlympiadBench | 56.4 | Pass@1 (%) | 85.7 |
| Token Rewards Evaluation - GaoKao2023en | 78.3 | Pass@1 (%) | 78.6 |
| GeoBenchLLM - PPNL Single-Goal | 83 | Optimal route ratio (%) | 64.3 |
| GeoBenchLLM - GridRoute | 66 | Optimal route ratio (%) | 57.1 |
| GeoBenchLLM - PPNL Multi-Goal | 23 | Optimal route ratio (%) | 57.1 |
| GeoBenchLLM - SpatialEvalLLM | 15 | Accuracy (%) | 57.1 |
| GeoBenchLLM - StepGame | 36 | Accuracy (%) | 57.1 |
Interactive version: theaggregate.ai/model?slug=qwen-3-1-7b-thinking · How It Works · Data refreshed daily, snapshot 2026-10-20.