Kaggle MultiLoKo: leaderboard
Metric: Exact Match (%). Source: www.kaggle.com. 48 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 57.81 |
| 2 | Gemini 3.5 Flash | 52.16 |
| 3 | Gemini 3 Flash (Preview) | 49.94 |
| 4 | Gemini 2.5 Pro (Preview 06-05) | 46.75 |
| 5 | GPT-5 | 40.89 |
| 6 | O3 | 40.88 |
| 7 | O1 | 39.78 |
| 8 | Gemini 3.1 Flash Lite (Preview) | 37.64 |
| 9 | Gemini 2.5 Flash (Preview 05-20) | 35.61 |
| 10 | GPT-5.4 | 33.41 |
| 11 | GPT-4.1 | 31.95 |
| 12 | GPT-4o (2024-08-06) | 31.61 |
| 13 | Gemini 2.0 Flash | 31.38 |
| 14 | Gemini 1.5 Pro (002) | 27.69 |
| 15 | Claude 3.7 Sonnet | 27.37 |
Interactive version: theaggregate.ai/benchmark?slug=kaggle-multiloko · How It Works · Data refreshed daily, snapshot 2026-09-19.