Kaggle MultiLoKo - French: leaderboard

Metric: Exact Match (%). Source: www.kaggle.com. 48 models tracked.

Top models

#ModelScore
1Gemini 3.1 Pro (Preview)74.19
2Gemini 3.5 Flash66.4
3Gemini 3 Flash (Preview)66
4Gemini 2.5 Pro (Preview 06-05)60
5O354.8
6GPT-553.6
7O151.6
8Gemini 2.5 Flash (Preview 05-20)47.6
9Gemini 3.1 Flash Lite (Preview)42.8
10DeepSeek R1 052840.8
11Claude 3.5 Sonnet (20241022)40
12GPT-4o (2024-08-06)39.6
13GPT-4.139.2
14Claude 3.7 Sonnet38.4
15Claude Opus 438

Interactive version: theaggregate.ai/benchmark?slug=kaggle-multiloko-french · How It Works · Data refreshed daily, snapshot 2026-09-19.