A2RBench: leaderboard
Metric: Total Acc (self-reported). Source: benchmarklist.com. 14 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5 | 39 |
| 2 | GPT-5 Mini | 36.9 |
| 3 | O4 Mini | 33.7 |
| 4 | Claude Sonnet 4.5 | 30.6 |
| 5 | GPT-5.2 | 28.3 |
| 6 | Gemini 2.5 Flash | 27.7 |
| 7 | GLM-4.6 | 21 |
| 8 | DeepSeek V3.2 | 20.8 |
| 9 | Qwen 3 32B | 19.3 |
| 10 | GPT-4o Mini | 16.1 |
| 11 | Qwen 3 14B | 14.2 |
| 12 | Qwen 3 8B | 0 |
Interactive version: theaggregate.ai/benchmark?slug=a2rbench · How It Works · Data refreshed daily, snapshot 2026-09-05.