Kaggle MultiLoKo - Italian Task v6: leaderboard
Meta MultiLoKo native Kaggle revisions: 30 languages at task v6 and German at v2, with 250 items per language where evaluated. Average v2 requires numeric results for all 31 tasks. Language boards retain their reported rows. These revisions are distinct from the historical EM tasks. Urdu retains the existing reviewed scoring hold.
Metric: Exact Match (%). Source: www.kaggle.com. Saturation forecast: Around July 2027. 16 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 71.6 | #60 |
| 2 | Gemini 3.8 Flash | 69.6 | #23 |
| 3 | GPT-6 Astra | 67.2 | #6 |
| 4 | GPT-6.1 Sol | 66 | #7 |
| 5 | Gemini 4 Argon | 65.6 | #8 |
| 6 | GPT-6 Luna | 56 | #61 |
| 7 | Qwen 3.8 Max | 48 | #32 |
| 8 | Claude Opus 5.5 | 47.2 | #5 |
| 9 | Kimi K3 | 46.8 | #33 |
| 10 | Grok 4.7 | 44.4 | #26 |
| 11 | Claude Sonnet 5.5 | 41.6 | #11 |
| 12 | GLM-5 | 34 | #146 |
| 13 | GLM-5.3 Flash | 31.2 | #73 |
| 14 | Claude Sonnet 5 | 30.8 | #62 |
| 15 | Claude Opus 4.8 | 25.6 | #37 |
No result here: #9 Claude Fable 5.1, #10 Muse Wonder Full.
Interactive version: theaggregate.ai/benchmark?slug=kaggle-multiloko-italian-task-v6 · How It Works · Data refreshed daily, snapshot 2026-10-25.