GraphVerse - Graph Edit Distance (Paired Images) - VGR-Score: leaderboard
Metric: Visual graph reasoning score (0-100; 0.7 process + 0.3 answer, GPT-5.1 verifier). Source: arxiv.org. 19 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3.5 35B A3B | 38.6 |
| 2 | Gemini 3 Pro | 35.1 |
| 3 | Kimi K2.5 (Thinking) | 34.5 |
| 4 | QVQ-72B-Preview | 32.7 |
| 5 | Gemini 3 Flash | 32.2 |
| 6 | GPT-5.2 | 30.5 |
| 7 | GLM-4.6V | 26.2 |
| 8 | O1 | 25.2 |
| 9 | Qwen 3.5 4B | 25.1 |
| 10 | Grok 4 Fast | 24.5 |
| 11 | InternVL3-8B | 6.8 |
Interactive version: theaggregate.ai/benchmark?slug=graphverse-graph-edit-distance-paired-images-vgr-score · How It Works · Data refreshed daily, snapshot 2026-09-19.