EuroEval French NLU - Allocine — leaderboard

Metric: Sentiment classification Score (%). Source: euroeval.com. 377 models tracked.

Top models

#ModelScore
1O3 (2025-04-16)98.19
2Ministral 3 14B97.8
3Bielik-11B-v2.3-Instruct97.49
4GPT-5 (High)97.33
5GPT-597.1
6GPT-5 (Minimal)97.1
7Ministral-3-14B-Instruct-251296.93
8Grok 4.1 Fast (Reasoning)96.79
9GPT-5.4 Mini (High)96.79
10Gemini 3 Pro (Preview)96.7
11Gemini 3 Flash (Preview) (Non-reasoning)96.7
12Ministral-3-8B-Instruct-251296.63
13Ministral 3 8B96.58
14Grok 396.47
15GPT-5 Nano (High)96.47

Interactive version: theaggregate.ai/benchmark?slug=euroeval-french-nlu-allocine · How the rankings work · Data refreshed daily, snapshot 2026-07-22.