RuleWorld - Parallel Multi-Rule (First-Order Logic): leaderboard
RuleWorld (2026) parallel multi-rule questions combining the conclusions of 2 to 8 independent rules, with all 1,000 rules of the pool written in first-order logic and placed in the prompt. Exact-match accuracy in percent, with partial credit per matching answer position; higher is better.
Metric: Exact-match accuracy (%; x100 of the 0-1 score; parallel multi-rule QA with 2 to 8 gold rules, mean of the seven rule counts, rules in first-order logic; all 1,000 abstract, non-commonsense procedural rules of the pool prepended to the prompt (full-context prompting); exact match on the boxed answer list, partial credit per matching position; 10 questions per task type and difficulty level for each of five seeds; temperature 0 (Qwen3-32B 0.6)). Source: arxiv.org. Saturation forecast: Estimated already saturated. 10 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Sonnet 4.6 | 91.17 | #158 |
| 2 | GPT-5.5 (2026-04-23) | 85.93 | #33 |
| 3 | DeepSeek V3.2 (Non-reasoning) | 78.21 | #403 (DeepSeek V3.2) |
| 4 | GPT-5.1 (2025-11-13) | 75.76 | #184 |
| 5 | Qwen 2.5 72B Instruct | 59.45 | #1093 |
| 6 | Llama 3.1 70B Instruct | 55.71 | #1541 |
| 7 | Qwen 3 32B | 53.26 | #960 |
| 8 | Qwen 2.5 32B Instruct | 50.38 | #1216 |
| 9 | Llama 3.3 70B Instruct | 45.55 | #1353 |
| 10 | Qwen 2.5 14B Instruct | 44.86 | #1745 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=ruleworld-parallel-multi-rule-first-order-logic · How It Works · Data refreshed daily, snapshot 2026-11-12.