LA Leaderboard - caBREU Catalan: leaderboard
Summarisation of Catalan news articles (caBREU). From the LA Leaderboard, which evaluates models on Spanish, Catalan, Basque and Galician tasks with the lm-evaluation-harness; sacreBLEU (0-100), higher is better. Each model result file is a separate run.
Metric: BLEU. Source: huggingface.co. Saturation forecast: Not forecast. 69 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | EuroLLM-9B | 19.55 | #1238 |
| 2 | Mistral-7B-v0.3 | 19.36 | #1321 |
| 3 | occiglot-7B-eu5 | 18.6 | |
| 4 | aya-expanse-8B | 18.55 | #914 |
| 5 | salamandra-7B | 18.35 | #1446 |
| 6 | Gemma 2 9B | 18.07 | #924 |
| 7 | Llama 3.1 8B | 17.42 | #1130 |
| 8 | Gemma 2 2B | 17 | #1466 |
| 9 | Llama 3.2 3B | 16.88 | #1376 |
| 10 | Yi-1.5-9B | 16.45 | #1215 |
| 11 | SmolLM3-3B-Base | 15.54 | #1260 |
| 12 | SmolLM2-1.7B | 15 | #1524 |
| 13 | occiglot-7B-eu5-instruct | 14.37 | |
| 14 | EuroLLM-1.7B-Instruct | 14.13 | #1554 |
| 15 | salamandra-2B-instruct | 13.45 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=la-leaderboard-cabreu-catalan · How It Works · Data refreshed daily, snapshot 2026-11-05.