EuroEval Spanish NLU - Sentiment Headlines ES — leaderboard

Metric: Sentiment classification Score (%). Source: euroeval.com. 414 models tracked.

Top models

#ModelScore
1Llama 3.1 70B Instruct53.79
2GPT-5.4 Nano (Medium)53.49
3Llama 3.3 70B Instruct53.31
4Gemma 4 31B (IT)53.02
5Mistral Small 3.152.99
6Gemini 2.5 Flash (Non-reasoning)52.95
7Claude Haiku 4.5 (20251001)52.68
8Gemma 3 27B (IT)52.65
9Magistral Small52.62
10Mistral Small 3.252.6
11Qwen 3 235B A22B 2507 Instruct52.26
12Apertus-70B-Instruct-250952.1
13Mistral Small 351.95
14Llama 4 Scout Instruct51.84
15Qwen 3 Next 80B A3B Instruct51.8

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