PetriBench - L4-Liveness (Hard): leaderboard
Exact-match accuracy on PetriBench's L4-liveness questions at the Hard structural-complexity level, one of six task families over Petri-net state spaces. Reasoning effort is part of the row identity. Higher is better. Source: arxiv.org/abs/2609.19883.
Metric: Exact-Match Accuracy (%). Source: arxiv.org. Saturation forecast: Around February 2027. 36 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 5 (High) | 93.5 | #16 (Claude Opus 5) |
| 2 | Claude Opus 5 (xHigh) | 92.5 | #16 (Claude Opus 5) |
| 3 | GPT-5.6 Sol (xHigh) | 91 | #22 (GPT-5.6 Sol) |
| 4 | Grok 4.6 (Medium) | 84 | #29 (Grok 4.6) |
| 5 | Gemini 3.1 Pro (Preview) (High) | 80 | #64 (Gemini 3.1 Pro (Preview)) |
| 6 | Grok 4.6 (High) | 80 | #29 (Grok 4.6) |
| 7 | Grok 4.6 (xHigh) | 79 | #29 (Grok 4.6) |
| 8 | Claude Opus 5 (Medium) | 78.5 | #16 (Claude Opus 5) |
| 9 | Gemini 3.8 Flash (High) | 78 | #30 (Gemini 3.8 Flash) |
| 10 | GPT-5.6 Terra (xHigh) | 77 | #40 (GPT-5.6 Terra) |
| 11 | GPT-5.6 Sol (High) | 74.5 | #22 (GPT-5.6 Sol) |
| 12 | Grok 4.5 (High) | 74 | #54 (Grok 4.5) |
| 13 | Grok 4.5 (Medium) | 73.5 | #54 (Grok 4.5) |
| 14 | GPT-5.6 Luna (xHigh) | 61.5 | #79 (GPT-5.6 Luna) |
| 15 | GPT-5.6 Sol (Medium) | 58 | #22 (GPT-5.6 Sol) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=petribench-l4-liveness-hard · How It Works · Data refreshed daily, snapshot 2026-11-05.