LatamBoard - Translation Score — leaderboard
Metric: Score (%). Source: huggingface.co. 35 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Hermes-4-14B | 48.25 |
| 2 | aya-expanse-8B | 48.24 |
| 3 | Apertus-8B-Instruct-2509 | 47.99 |
| 4 | Hermes-3-Llama-3.1-8B | 46.55 |
| 5 | Phi-3-medium-4k-instruct | 46.24 |
| 6 | Qwen 3 VL 8B Instruct | 45.11 |
| 7 | Granite 4.0 H 1B | 44.15 |
| 8 | Phi-4 Multimodal Instruct | 44.09 |
| 9 | SOLAR-10.7B-Instruct-v1.0 | 42.47 |
| 10 | zephyr-7B-beta | 42.08 |
| 11 | Ministral-8B-Instruct-2410 | 37.28 |
| 12 | Nemotron-Mini-4B-Instruct | 32.62 |
| 13 | Gemma 3n E4B (IT) | 30.23 |
| 14 | gemma-3-12B-pt | 28.31 |
| 15 | GPT-OSS-20B | 28.24 |
Interactive version: theaggregate.ai/benchmark?slug=latamboard-translation-score · How the rankings work · Data refreshed daily, snapshot 2026-07-22.