EuroEval Spanish NLU - Sentiment Headlines ES: leaderboard
The "Sentiment Headlines ES" subtask of EuroEval Spanish NLU. Spanish-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: Around November 2027. 414 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | cogito-v1-preview-qwen-14B | 53.82 | |
| 2 | Llama 3.1 70B Instruct | 53.79 | #574 |
| 3 | Llama 3.3 70B Instruct | 53.31 | #541 |
| 4 | Gemma 4 31B (IT) | 53.02 | #179 |
| 5 | Mistral Small 3.1 | 52.99 | #625 |
| 6 | Gemini 2.5 Flash (Non-reasoning) | 52.95 | #254 (Gemini 2.5 Flash) |
| 7 | Qwen 3 32B (Non-reasoning) | 52.82 | #449 (Qwen 3 32B) |
| 8 | Claude Haiku 4.5 (20251001) | 52.68 | #263 |
| 9 | Gemma 3 27B (IT) | 52.65 | #529 |
| 10 | Magistral Small 1.2 | 52.62 | #605 |
| 11 | Mistral Small 3.2 | 52.6 | #608 |
| 12 | Voxtral-Small-24B-2507 | 52.43 | #594 |
| 13 | Qwen 3 235B A22B 2507 Instruct FP8 | 52.26 | #231 |
| 14 | Apertus-70B-Instruct-2509 | 52.1 | #822 |
| 15 | Mistral Small 3 | 51.95 | #631 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #6 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=euroeval-spanish-nlu-sentiment-headlines-es · How It Works · Data refreshed daily, snapshot 2026-11-06.