LatamBoard - Spanish XNLI: leaderboard
Metric: Score (%). Source: huggingface.co. 35 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Apertus-8B-Instruct-2509 | 50.76 |
| 2 | Llama 3.1 8B Instruct | 50.24 |
| 3 | Hermes-4-14B | 49.44 |
| 4 | Ministral-8B-Instruct-2410 | 48.71 |
| 5 | Qwen 3 VL 8B Instruct | 48.43 |
| 6 | Nemotron-Mini-4B-Instruct | 48.23 |
| 7 | Granite 3.3 8B Instruct | 48.07 |
| 8 | Mistral-NeMo-Minitron-8B-Instruct | 47.83 |
| 9 | Hermes-3-Llama-3.1-8B | 47.67 |
| 10 | Phi-3-medium-4k-instruct | 47.55 |
| 11 | Phi-4 Mini Instruct | 47.47 |
| 12 | Qwen 3 VL 4B Instruct | 47.35 |
| 13 | Phi-4 Multimodal Instruct | 47.07 |
| 14 | aya-expanse-8B | 46.75 |
| 15 | Qwen 3 4B 2507 Instruct | 46.18 |
Interactive version: theaggregate.ai/benchmark?slug=latamboard-spanish-xnli · How It Works · Data refreshed daily, snapshot 2026-09-05.