LatamBoard - Spanish PAWS: leaderboard
The "Spanish PAWS" subtask of LatamBoard. Leaderboard for model performance on Latin American and Iberian language tasks, including Spanish and Portuguese understanding, TELEIA Spanish exams, FLORES/OPUS translation, and structured or image extraction.
Metric: Score (%). Source: huggingface.co. Saturation forecast: Not forecast. 35 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Hermes-4-14B | 66.35 | #751 |
| 2 | Llama 3.1 8B Instruct | 64.85 | #1035 |
| 3 | SOLAR-10.7B-Instruct-v1.0 | 63.8 | #1075 |
| 4 | Yi-1.5-9B Chat | 63 | #1031 |
| 5 | Mistral-NeMo-Minitron-8B-Instruct | 63 | #907 |
| 6 | Qwen 3 VL 8B Instruct | 62.55 | #420 |
| 7 | Ministral-8B-Instruct-2410 | 62 | #1094 |
| 8 | Yi-1.5-6B-Chat | 61.5 | #1259 |
| 9 | Qwen 3 4B 2507 Instruct | 61.45 | #762 |
| 10 | Apertus-8B-Instruct-2509 | 61.2 | #1061 |
| 11 | Gemma 3n E4B (IT) | 61.15 | #749 |
| 12 | DeepSeek R1 Distill Llama 8B | 60.95 | #1291 |
| 13 | aya-expanse-8B | 60.9 | #931 |
| 14 | Qwen 3 VL 4B Instruct | 60.65 | #515 |
| 15 | AFM-4.5B | 60.35 | #1372 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=latamboard-spanish-paws · How It Works · Data refreshed daily, snapshot 2026-10-20.