LA Leaderboard - Fake News Spanish: leaderboard
Classifying Spanish news articles as real or fake. From the LA Leaderboard, which evaluates models on Spanish, Catalan, Basque and Galician tasks with the lm-evaluation-harness; accuracy (%), higher is better. Each model result file is a separate run.
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 69 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 2.5 7B Instruct | 84.27 | #863 |
| 2 | Mistral 7B Instruct (v0.3) | 82.17 | #1255 |
| 3 | Gemma 2 9B (IT) | 81.47 | #795 |
| 4 | Llama 3.2 3B Instruct | 80.77 | #1327 |
| 5 | Llama 3.1 8B Instruct | 80.07 | #1017 |
| 6 | Qwen 2.5 3B Instruct | 79.37 | #1159 |
| 7 | Qwen3-8B-Base | 77.1 | #626 |
| 8 | OLMo-2-1124-7B-Instruct | 73.95 | #1343 |
| 9 | Gemma 3 4B (IT) | 73.25 | #997 |
| 10 | Qwen 2.5 7B | 71.33 | #895 |
| 11 | Yi-1.5-9B Chat | 70.45 | #1032 |
| 12 | EuroLLM-9B-Instruct | 69.06 | #1047 |
| 13 | SmolLM3-3B | 69.06 | #1239 |
| 14 | occiglot-7B-eu5-instruct | 67.48 | |
| 15 | Gemma 2 2B (IT) | 64.86 | #1328 |
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-fake-news-spanish · How It Works · Data refreshed daily, snapshot 2026-11-05.