Kaggle MultiLoKo - German: leaderboard
Metric: Exact Match (%). Source: www.kaggle.com. 48 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 64.9 |
| 2 | Gemini 3 Flash (Preview) | 58 |
| 3 | Gemini 3.5 Flash | 56.8 |
| 4 | Gemini 2.5 Pro (Preview 06-05) | 52.4 |
| 5 | GPT-5 | 48.8 |
| 6 | O3 | 45.6 |
| 7 | O1 | 43.6 |
| 8 | Gemini 3.1 Flash Lite (Preview) | 42.8 |
| 9 | Claude 3.7 Sonnet | 41.2 |
| 10 | Claude 3.5 Sonnet (20241022) | 39.2 |
| 11 | GPT-5.4 | 39.2 |
| 12 | Gemini 2.5 Flash (Preview 05-20) | 37.2 |
| 13 | GPT-4.1 | 35.6 |
| 14 | Gemini 2.0 Flash | 34.4 |
| 15 | Claude Opus 4 | 34.4 |
Interactive version: theaggregate.ai/benchmark?slug=kaggle-multiloko-german · How It Works · Data refreshed daily, snapshot 2026-09-19.