Knossos (Edge F1): leaderboard
Knossos (2026) evaluates complete diagram-to-graph topology extraction: from a rendered diagram, the model must output every entity and every relation between them. Its 1,200-diagram test set spans six symbolically generated domains (food web, network, workflow, natural process, circuit, map route) with exact ground-truth topology. The board is the Knossos Avg. column of the paper's Table 1: Edge F1 (%) macro-averaged over the six domains, with each VLM baseline prompted for direct full-graph generation in one response. Node extraction is near ceiling for current models (97%+ Node F1 for the strongest), so Edge F1 carries the signal.
Metric: Edge F1 (%), macro-averaged over six diagram domains (1,200 test diagrams), direct full-graph generation. Source: arxiv.org. Saturation forecast: Around February 2027. 10 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.6 Sol | 81.1 | #37 |
| 2 | Claude Opus 5 | 70.4 | #26 |
| 3 | Qwen 3 VL 32B | 54.7 | #622 |
| 4 | Qwen 3 VL 8B | 46.1 | #917 |
| 5 | GPT-4o | 41.2 | #751 |
| 6 | InternVL3.5-8B | 40.7 | #1604 |
| 7 | MiMo-VL-7B-RL | 40.7 | #1155 |
| 8 | Keye-VL-1.5-8B | 39 | #1720 |
| 9 | GLM-4.6V-Flash | 38.9 | #1040 |
| 10 | Ovis2.5-9B | 38.3 | #1225 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=knossos-edge-f1 · How It Works · Data refreshed daily, snapshot 2026-11-21.