Time Puzzles - Six-condition Exact Match: leaderboard
Metric: Exact-match Accuracy (%). Source: arxiv.org. 13 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5 | 55.33 |
| 2 | DeepSeek V3.2 (Thinking) | 36.67 |
| 3 | GPT-4.1 | 35.17 |
| 4 | DeepSeek V3.2 (Non-reasoning) | 30.5 |
| 5 | GPT-5 Nano | 15.83 |
| 6 | Qwen 3 14B | 13.17 |
| 7 | GPT-OSS-20B | 10.83 |
| 8 | Qwen 3 8B | 10.83 |
| 9 | GPT-4.1 Nano | 9.33 |
| 10 | Qwen 3 4B 2507 (Thinking) | 5.67 |
| 11 | Ministral-3-8B-Instruct-2512 | 3.5 |
| 12 | Qwen 3 4B 2507 Instruct | 3 |
Interactive version: theaggregate.ai/benchmark?slug=time-puzzles-six-condition-exact-match · How It Works · Data refreshed daily, snapshot 2026-09-19.