Kaggle MultiLoKo - Korean Task v7: leaderboard
Native model scores in the source’s stated configuration. Metric: Exact Match (%). Source: https://www.kaggle.com/benchmarks/metaresearch/multiloko
Metric: Exact Match (%). Source: www.kaggle.com. Saturation forecast: Around January 2028. 16 models tracked. Updated 17 Nov 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6.1 Sol | 57.2 | #12 |
| 2 | GPT-6 Astra | 56 | #8 |
| 3 | Gemini 3.1 Pro (Preview) | 55.2 | #99 |
| 4 | GPT-6 Luna | 48 | #105 |
| 5 | Gemini 3.8 Flash | 46.8 | #39 |
| 6 | Qwen 3.8 Max | 43.2 | #64 |
| 7 | Grok 4.7 | 41.6 | #48 |
| 8 | Gemini 4 Argon | 35.2 | #14 |
| 9 | GLM-5 | 24.8 | #267 |
| 10 | Claude Opus 5.5 | 23.2 | #7 |
| 11 | Claude Sonnet 5.5 | 19.2 | #21 |
| 12 | Gemma 4 31B | 18.4 | #385 |
| 13 | Kimi K3 | 13.2 | #59 |
| 14 | Claude Sonnet 5 | 12.8 | #118 |
| 15 | GLM-5.3 Flash | 10.4 | #136 |
No result here: #1 Claude Fable 5.2, #2 GPT-6.2 Astra, #3 Gemini 4 Carbon.
Interactive version: theaggregate.ai/benchmark?slug=kaggle-multiloko-korean-task-v7 · How It Works · Data refreshed daily, snapshot 2026-11-17.