LegendBench - Guide Reading: leaderboard
Metric: Guide-reading accuracy (%; legend-parsing questions: extract the mapping the legend makes explicit; LegendBench snapshot of 26,689 legend-centric chart questions generated from 900 base charts with counterfactual legend variants; the same prompt template and hyperparameters for every model). Source: arxiv.org. Saturation forecast: Around December 2026. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.6 Sol | 91.12 |
| 2 | Qwen 3.5 27B | 86.19 |
| 3 | Gemini 2.5 Flash | 79.81 |
| 4 | GLM-4.6V | 78.74 |
| 5 | GPT-4.1 Mini | 76.82 |
Interactive version: theaggregate.ai/benchmark?slug=legendbench-guide-reading · How It Works · Data refreshed daily, snapshot 2026-09-26.