Kaggle MultiLoKo - English: leaderboard
Metric: Exact Match (%). Source: www.kaggle.com. 48 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 74.8 |
| 2 | GPT-5 | 71.2 |
| 3 | O3 | 69.6 |
| 4 | O1 | 66.4 |
| 5 | Gemini 2.5 Pro (Preview 06-05) | 65.6 |
| 6 | Claude Opus 4.8 | 61.6 |
| 7 | GPT-4.1 | 59.6 |
| 8 | Gemini 3.1 Flash Lite (Preview) | 57.2 |
| 9 | Gemini 2.5 Flash (Preview 05-20) | 56.8 |
| 10 | GPT-4o (2024-08-06) | 56.4 |
| 11 | GPT-5.4 | 55.6 |
| 12 | O4 Mini | 54.8 |
| 13 | Gemini 3.5 Flash | 54 |
| 14 | Claude 3.5 Sonnet (20241022) | 52.4 |
| 15 | Grok 3 | 50.4 |
Interactive version: theaggregate.ai/benchmark?slug=kaggle-multiloko-english · How It Works · Data refreshed daily, snapshot 2026-09-19.