LatamBoard - Spanish XNLI: leaderboard
The "Spanish XNLI" 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 | Apertus-8B-Instruct-2509 | 50.76 | #1061 |
| 2 | Llama 3.1 8B Instruct | 50.24 | #1035 |
| 3 | Hermes-4-14B | 49.44 | #751 |
| 4 | Ministral-8B-Instruct-2410 | 48.71 | #1094 |
| 5 | Qwen 3 VL 8B Instruct | 48.43 | #420 |
| 6 | Granite 3.3 8B Instruct | 48.07 | #825 |
| 7 | Mistral-NeMo-Minitron-8B-Instruct | 47.83 | #907 |
| 8 | Hermes-3-Llama-3.1-8B | 47.67 | #997 |
| 9 | Phi-3-medium-4k-instruct | 47.55 | #952 |
| 10 | Phi-4 Mini Instruct | 47.47 | #1245 |
| 11 | Qwen 3 VL 4B Instruct | 47.35 | #515 |
| 12 | Phi-4 Multimodal Instruct | 47.07 | #906 |
| 13 | aya-expanse-8B | 46.75 | #931 |
| 14 | Qwen 3 4B 2507 Instruct | 46.18 | #762 |
| 15 | DeepSeek R1 Distill Llama 8B | 45.5 | #1291 |
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-xnli · How It Works · Data refreshed daily, snapshot 2026-10-20.