Kaggle MultiLoKo - German Task v3: 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 August 2027. 16 models tracked. Updated 17 Nov 2026.

Top models

#ModelScoreOverall rank
1Gemini 3.1 Pro (Preview)65.2#99
2GPT-6.1 Sol63.6#12
3GPT-6 Astra60.8#8
4Gemini 3.8 Flash58.4#39
5Gemini 4 Argon54#14
6Grok 4.746.4#48
7GPT-6 Luna44.4#105
8Qwen 3.8 Max41.2#64
9Claude Opus 5.540.4#7
10GLM-536#267
11Kimi K335.2#59
12Claude Sonnet 5.534#21
13Claude Opus 4.830.4#62
14Claude Sonnet 529.2#118
15GLM-5.3 Flash24#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-german-task-v3 · How It Works · Data refreshed daily, snapshot 2026-11-17.