PetriBench - ReachableMarkings (Hard): leaderboard
Exact-match accuracy on PetriBench's reachable-markings 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 | GPT-5.6 Sol (xHigh) | 86.2 | #22 (GPT-5.6 Sol) |
| 2 | Claude Opus 5 (High) | 82 | #16 (Claude Opus 5) |
| 3 | Claude Opus 5 (xHigh) | 72.2 | #16 (Claude Opus 5) |
| 4 | Claude Opus 5 (Medium) | 67 | #16 (Claude Opus 5) |
| 5 | GPT-5.6 Sol (High) | 57.8 | #22 (GPT-5.6 Sol) |
| 6 | GPT-5.6 Terra (xHigh) | 50.5 | #40 (GPT-5.6 Terra) |
| 7 | Gemini 3.8 Flash (High) | 29 | #30 (Gemini 3.8 Flash) |
| 8 | Grok 4.6 (xHigh) | 27 | #29 (Grok 4.6) |
| 9 | Grok 4.6 (High) | 22.8 | #29 (Grok 4.6) |
| 10 | GPT-5.6 Luna (xHigh) | 22.2 | #79 (GPT-5.6 Luna) |
| 11 | Grok 4.6 (Medium) | 21.8 | #29 (Grok 4.6) |
| 12 | Gemini 3.1 Pro (Preview) (High) | 13.5 | #64 (Gemini 3.1 Pro (Preview)) |
| 13 | Grok 4.5 (High) | 13.2 | #54 (Grok 4.5) |
| 14 | GPT-5.6 Sol (Medium) | 12.8 | #22 (GPT-5.6 Sol) |
| 15 | Grok 4.5 (Medium) | 12 | #54 (Grok 4.5) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=petribench-reachablemarkings-hard · How It Works · Data refreshed daily, snapshot 2026-11-05.