LatamBoard - ASSIN2 RTE: leaderboard
The "ASSIN2 RTE" 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: Estimated already saturated. 35 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Hermes-4-14B | 94.36 | #751 |
| 2 | Qwen 3 VL 8B Instruct | 94.2 | #420 |
| 3 | Qwen 3 VL 4B Instruct | 94.08 | #515 |
| 4 | Phi-3-medium-4k-instruct | 93.43 | #952 |
| 5 | Qwen 3 4B 2507 Instruct | 92.64 | #762 |
| 6 | Gemma 3n E4B (IT) | 92.6 | #749 |
| 7 | SOLAR-10.7B-Instruct-v1.0 | 92.52 | #1075 |
| 8 | Gemma 3n E2B (IT) | 92.4 | #942 |
| 9 | Ministral-8B-Instruct-2410 | 92.06 | #1094 |
| 10 | gemma-3-12B-pt | 91.93 | #686 |
| 11 | Granite 3.3 8B Instruct | 91.91 | #825 |
| 12 | Apertus-8B-Instruct-2509 | 91.75 | #1061 |
| 13 | Hermes-3-Llama-3.1-8B | 91.33 | #997 |
| 14 | aya-vision-8B | 91.06 | #1107 |
| 15 | Llama 3.1 8B Instruct | 90.74 | #1035 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=latamboard-assin2-rte · How It Works · Data refreshed daily, snapshot 2026-10-20.