LatamBoard - BLUEX: leaderboard
The "BLUEX" 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: Around 2029. 35 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Hermes-4-14B | 74.27 | #751 |
| 2 | Qwen 3 VL 8B Instruct | 72.88 | #420 |
| 3 | gemma-3-12B-pt | 68.01 | #686 |
| 4 | Qwen 3 VL 4B Instruct | 67.59 | #515 |
| 5 | Qwen 3 4B 2507 Instruct | 67.32 | #762 |
| 6 | Phi-3-medium-4k-instruct | 64.53 | #952 |
| 7 | Mistral-NeMo-Minitron-8B-Instruct | 61.34 | #907 |
| 8 | Gemma 3n E4B (IT) | 60.5 | #749 |
| 9 | Llama 3.1 8B Instruct | 60.36 | #1035 |
| 10 | Ministral-8B-Instruct-2410 | 59.81 | #1094 |
| 11 | SOLAR-10.7B-Instruct-v1.0 | 58.28 | #1075 |
| 12 | Granite 3.3 8B Instruct | 57.86 | #825 |
| 13 | Hermes-3-Llama-3.1-8B | 57.3 | #997 |
| 14 | Granite 4.0 H 1B | 56.88 | #1191 |
| 15 | Gemma 3n E2B (IT) | 55.91 | #942 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=latamboard-bluex · How It Works · Data refreshed daily, snapshot 2026-10-20.