LA Leaderboard - NoticIA: leaderboard
Summarising Spanish clickbait news articles in one sentence that resolves the headline (NoticIA). From the LA Leaderboard, which evaluates models on Spanish, Catalan, Basque and Galician tasks with the lm-evaluation-harness; ROUGE-1 (%), higher is better. Each model result file is a separate run.
Metric: ROUGE-1 (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 69 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemma 2 9B (IT) | 34.11 | #795 |
| 2 | Qwen 2.5 7B Instruct | 31.47 | #863 |
| 3 | Llama 3.1 8B Instruct | 28.71 | #1017 |
| 4 | Qwen 2.5 3B Instruct | 28.36 | #1159 |
| 5 | Mistral 7B Instruct (v0.3) | 27.44 | #1255 |
| 6 | EuroLLM-9B-Instruct | 26.8 | #1047 |
| 7 | Gemma 2 2B (IT) | 25.7 | #1328 |
| 8 | Llama 3.2 3B Instruct | 25.4 | #1327 |
| 9 | Gemma 3 4B (IT) | 23.93 | #997 |
| 10 | OLMo-2-1124-7B-Instruct | 23.29 | #1343 |
| 11 | occiglot-7B-eu5-instruct | 23.08 | |
| 12 | Qwen 2.5 1.5B Instruct | 22.07 | #1393 |
| 13 | Yi-1.5-9B Chat | 21.49 | #1032 |
| 14 | Teuken-7B-Instruct-v0.4 | 20.37 | #1408 |
| 15 | salamandra-7B-instruct | 19.38 | #1392 |
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-noticia · How It Works · Data refreshed daily, snapshot 2026-11-05.