LA Leaderboard - XQuAD Spanish: leaderboard
XQuAD extractive question answering in Spanish. From the LA Leaderboard, which evaluates models on Spanish, Catalan, Basque and Galician tasks with the lm-evaluation-harness; F1 (%), higher is better. Each model result file is a separate run.
Metric: F1 (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 69 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 2.5 7B | 80.05 | #895 |
| 2 | Yi-1.5-9B | 79.5 | #1215 |
| 3 | aya-expanse-8B | 77.98 | #914 |
| 4 | Qwen 2.5 3B | 77 | #1177 |
| 5 | EuroLLM-9B | 76.92 | #1238 |
| 6 | OLMo-2-1124-7B | 76.04 | #1381 |
| 7 | Gemma 2 9B | 75.33 | #924 |
| 8 | Llama 3.1 8B | 74.38 | #1130 |
| 9 | Mistral-7B-v0.3 | 72.56 | #1321 |
| 10 | Qwen 2.5 1.5B | 72.55 | #1412 |
| 11 | occiglot-7B-eu5 | 72.09 | |
| 12 | gemma-3-4B-pt Base | 72.03 | #1110 |
| 13 | Gemma 3 4B (IT) | 70.67 | #997 |
| 14 | Llama 3.1 8B Instruct | 69.94 | #1017 |
| 15 | SmolLM3-3B-Base | 69.8 | #1260 |
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-xquad-spanish · How It Works · Data refreshed daily, snapshot 2026-11-05.