GMP - Dynamic Rules (Coding Platform Discussion): leaderboard
Metric: F1 (x100) of the unsafe judgement under GMP Task B rule set 4, Coding Platform Discussion (delayed, anonymous; 2,000 posts judged against that context's listed prohibitions, everything else permitted), zero-shot prompt with JSON output; higher is better. Source: arxiv.org. Saturation forecast: Around January 2028. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Sonnet 4 | 47.88 | #194 |
| 2 | DeepSeek V3.1 (Non-reasoning) | 47.29 | #260 (DeepSeek V3.1) |
| 3 | Gemini 2.5 Flash | 46.94 | #237 |
| 4 | Qwen 2.5 VL 72B Instruct | 44.47 | #364 |
| 5 | Qwen 3 30B A3B 2507 Instruct | 43.09 | #464 |
| 6 | GPT-4o Mini | 42.39 | #588 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=gmp-dynamic-rules-coding-platform-discussion · How It Works · Data refreshed daily, snapshot 2026-10-11.