HELM Image2Struct - Image2Latex Table: leaderboard
Metric: EMS. Source: crfm.stanford.edu. 20 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4o (2024-05-13) | 71.1 |
| 2 | GPT-4o (2024-08-06) | 69.84 |
| 3 | Gemini 1.5 Pro (002) | 62.58 |
| 4 | Claude 3 Opus (20240229) | 60.73 |
| 5 | Gemini 1.5 Flash (002) | 59.98 |
| 6 | Claude 3.5 Sonnet (20241022) | 57.13 |
| 7 | Claude 3.5 Sonnet (20240620) | 50.43 |
| 8 | Claude 3 Sonnet (20240229) | 45.11 |
| 9 | GPT-4o Mini (2024-07-18) | 2.69 |
Interactive version: theaggregate.ai/benchmark?slug=helm-image2struct-image2latex-table · How It Works · Data refreshed daily, snapshot 2026-09-08.