Kaggle MultiLoKo - Khmer: leaderboard
Metric: Exact Match (%). Source: www.kaggle.com. 48 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 35.25 |
| 2 | Gemini 3.5 Flash | 32.4 |
| 3 | Gemini 3 Flash (Preview) | 31.6 |
| 4 | Gemini 3.1 Flash Lite (Preview) | 26 |
| 5 | Gemini 2.5 Pro (Preview 06-05) | 24 |
| 6 | Gemini 2.5 Flash (Preview 05-20) | 23.2 |
| 7 | GPT-5 | 19.2 |
| 8 | Grok 3 | 19.2 |
| 9 | Gemini 2.0 Flash | 18.8 |
| 10 | Gemini 1.5 Pro (002) | 18 |
| 11 | Claude Sonnet 4.6 | 16.8 |
| 12 | GPT-5.4 | 16 |
| 13 | Gemma 4 31B | 16 |
| 14 | O1 | 15.6 |
| 15 | O3 | 15.2 |
Interactive version: theaggregate.ai/benchmark?slug=kaggle-multiloko-khmer · How It Works · Data refreshed daily, snapshot 2026-09-19.