HyperGVL - Understanding: leaderboard
HyperGVL (2026) tests vision-language models on hypergraph understanding: counting vertices and hyperedges, degree- and order-specified counting and neighbor queries, each problem posed in 35 textual and visual representations. Average accuracy (%) over the six understanding tasks; higher is better.
Metric: Average accuracy (%) over the six understanding tasks (vertex, hyperedge, degree- and order-specified counting, neighbors, order-filtered neighbors), 7,000 questions per task (200 problems on synthetic and real-world hypergraphs, each posed in 35 textual, visual and combined hypergraph representations), accuracy (%) averaged over the representations, temperature 0.8, top-p 0.95; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 11 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Flash | 89.72 | #175 |
| 2 | Gemma 3 12B | 63.53 | #2059 |
| 3 | Qwen 3 VL 8B | 59.8 | #917 |
| 4 | InternVL3-8B | 44.96 | #1782 |
| 5 | Qwen 2.5 VL 7B Instruct | 42.39 | #1902 |
| 6 | GLM-4V-9B | 34.38 | #3142 |
| 7 | InternVL2.5-8B | 27.2 | #2370 |
| 8 | Molmo-7B-D-0924 | 18.6 | #7207 |
| 9 | DeepSeek-VL2 | 15.77 | #2927 |
| 10 | LLaVA-NeXT-7B | 11.44 | #5032 |
| 11 | LLaVA-v1.5-7B | 9.05 | #7630 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=hypergvl-understanding · How It Works · Data refreshed daily, snapshot 2026-11-21.