OrderProbe - Few-Shot: leaderboard
OrderProbe's exact-match recovery rate in percent (higher is better) when the prompt includes three solved examples, over the same scrambled four-character Chinese, Japanese and Korean expressions in all 23 non-identity permutations (Kang et al., ACL 2026).
Metric: Exact Recovery Rate (%; three-shot prompt). Source: arxiv.org. Saturation forecast: Around April 2028. 12 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 14B | 52.4 | #1235 |
| 2 | DeepSeek V3.2 | 49.1 | #403 |
| 3 | Qwen 3 8B | 39.3 | #1838 |
| 4 | DeepSeek R1 | 37.8 | #467 |
| 5 | Claude Haiku 4.5 | 35.8 | #568 |
| 6 | GPT-4o | 35.3 | #741 |
| 7 | Gemini 2.5 Flash | 34.6 | #452 |
| 8 | Gemma 3 4B (IT) | 28.8 | #3848 |
| 9 | Llama 2 7B Chat | 18.6 | #8862 |
| 10 | Llama 3 8B Instruct | 17.2 | #4652 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=orderprobe-few-shot · How It Works · Data refreshed daily, snapshot 2026-11-12.