LithoBench — leaderboard
Metric: MCQ$_{\mathrm{all}}$ (self-reported). Source: benchmarklist.com. 13 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3.5 9B | 98.25 |
| 2 | GLM-4.6V | 94.5 |
| 3 | GPT-5.4 | 93 |
| 4 | Qwen 3.5 397B A17B | 92.77 |
| 5 | Claude Sonnet 4.6 | 79.15 |
| 6 | Qwen 3.5 27B | 77.2 |
| 7 | Grok 4 | 76.05 |
Interactive version: theaggregate.ai/benchmark?slug=lithobench · How the rankings work · Data refreshed daily, snapshot 2026-07-22.