LA Leaderboard - Summarization Galician: leaderboard
Summarisation of Galician news articles. 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. 68 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Mistral-7B-v0.3 | 7.55 | #1321 |
| 2 | occiglot-7B-eu5 | 7.49 | |
| 3 | Llama 3.2 3B | 7.42 | #1376 |
| 4 | Yi-1.5-9B | 7.35 | #1215 |
| 5 | aya-expanse-8B | 7.26 | #914 |
| 6 | Gemma 2 9B | 7.02 | #924 |
| 7 | EuroLLM-9B | 6.89 | #1238 |
| 8 | SmolLM3-3B-Base | 6.88 | #1260 |
| 9 | SmolLM3-3B | 6.85 | #1239 |
| 10 | OLMo-2-1124-7B | 6.79 | #1381 |
| 11 | Llama 3.2 3B Instruct | 6.48 | #1327 |
| 12 | Llama 3.1 8B | 6.2 | #1130 |
| 13 | Llama 3.1 8B Instruct | 5.99 | #1017 |
| 14 | SmolLM2-1.7B | 5.92 | #1524 |
| 15 | Teuken-7B-instruct-commercial-v0.4 | 5.58 | #1421 |
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-summarization-galician · How It Works · Data refreshed daily, snapshot 2026-11-05.