LatamBoard - FaQuAD NLI: leaderboard
The "FaQuAD NLI" 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 | 82.81 | #751 |
| 2 | Qwen 3 VL 8B Instruct | 82.08 | #420 |
| 3 | Qwen 3 VL 4B Instruct | 81.37 | #515 |
| 4 | Qwen 3 4B 2507 Instruct | 81.19 | #762 |
| 5 | SOLAR-10.7B-Instruct-v1.0 | 78.57 | #1075 |
| 6 | aya-expanse-8B | 76.44 | #931 |
| 7 | Yi-1.5-9B Chat | 76.18 | #1031 |
| 8 | AFM-4.5B | 75.49 | #1372 |
| 9 | Apertus-8B-Instruct-2509 | 75.26 | #1061 |
| 10 | aya-vision-8B | 74.19 | #1107 |
| 11 | Phi-3-medium-4k-instruct | 73.42 | #952 |
| 12 | Granite 3.3 8B Instruct | 73.4 | #825 |
| 13 | Ministral-8B-Instruct-2410 | 71.47 | #1094 |
| 14 | Gemma 3n E4B (IT) | 71.19 | #749 |
| 15 | Hermes-3-Llama-3.1-8B | 71.08 | #997 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=latamboard-faquad-nli · How It Works · Data refreshed daily, snapshot 2026-10-20.