LatamBoard - FLORES Bidirectional: leaderboard
The "FLORES Bidirectional" 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 | 47.41 | #751 |
| 2 | aya-expanse-8B | 46.97 | #931 |
| 3 | Apertus-8B-Instruct-2509 | 46.76 | #1061 |
| 4 | Hermes-3-Llama-3.1-8B | 45.94 | #997 |
| 5 | Phi-3-medium-4k-instruct | 45.14 | #952 |
| 6 | Qwen 3 VL 8B Instruct | 43.86 | #420 |
| 7 | Granite 3.3 8B Instruct | 43.78 | #825 |
| 8 | Phi-4 Multimodal Instruct | 43.02 | #906 |
| 9 | Granite 4.0 H 1B | 42.71 | #1191 |
| 10 | SOLAR-10.7B-Instruct-v1.0 | 41.13 | #1075 |
| 11 | zephyr-7B-beta | 40.87 | #1330 |
| 12 | Ministral-8B-Instruct-2410 | 35.07 | #1094 |
| 13 | Gemma 3n E4B (IT) | 26.77 | #749 |
| 14 | gemma-3-12B-pt | 26.76 | #686 |
| 15 | GPT-OSS-20B | 26.74 | #518 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=latamboard-flores-bidirectional · How It Works · Data refreshed daily, snapshot 2026-10-20.