EuroEval French NLU - Allocine: leaderboard
The "Allocine" subtask of EuroEval French NLU. French-language EuroEval natural-language-understanding suite, separating NLU task performance from the broader all-task EuroEval aggregate.
Metric: Matthews correlation coefficient (%). Source: euroeval.com. Saturation forecast: Estimated already saturated. 377 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | O3 (2025-04-16) | 98.19 | #121 |
| 2 | Bielik-11B-v2.3-Instruct | 97.49 | #693 |
| 3 | GPT-5 (2025-08-07) (High) | 97.33 | |
| 4 | GPT-5 | 97.1 | #98 |
| 5 | Ministral-3-14B-Instruct-2512 | 96.93 | #615 |
| 6 | Grok 4.1 Fast (Reasoning) | 96.79 | #217 (Grok 4.1 Fast) |
| 7 | GPT-5.4 Mini (2026-03-17) (High) | 96.79 | #169 (GPT-5.4 Mini (2026-03-17)) |
| 8 | Gemini 3 Pro (Preview) | 96.7 | #66 |
| 9 | Gemini 3 Flash (Preview) (Non-reasoning) | 96.7 | #82 (Gemini 3 Flash (Preview)) |
| 10 | Gemini 3 Flash (Preview) (Thinking) | 96.7 | #82 (Gemini 3 Flash (Preview)) |
| 11 | Ministral-3-8B-Instruct-2512 | 96.63 | #661 |
| 12 | Grok 3 | 96.47 | #304 |
| 13 | GPT-5 Nano (2025-08-07) (High) | 96.47 | |
| 14 | Gemma 4 31B (IT) | 96.46 | #176 |
| 15 | Claude Sonnet 4.6 | 96.31 | #90 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=euroeval-french-nlu-allocine · How It Works · Data refreshed daily, snapshot 2026-10-20.