LatamBoard - Translation Score — leaderboard

Metric: Score (%). Source: huggingface.co. 35 models tracked.

Top models

#ModelScore
1Hermes-4-14B48.25
2aya-expanse-8B48.24
3Apertus-8B-Instruct-250947.99
4Hermes-3-Llama-3.1-8B46.55
5Phi-3-medium-4k-instruct46.24
6Qwen 3 VL 8B Instruct45.11
7Granite 4.0 H 1B44.15
8Phi-4 Multimodal Instruct44.09
9SOLAR-10.7B-Instruct-v1.042.47
10zephyr-7B-beta42.08
11Ministral-8B-Instruct-241037.28
12Nemotron-Mini-4B-Instruct32.62
13Gemma 3n E4B (IT)30.23
14gemma-3-12B-pt28.31
15GPT-OSS-20B28.24

Interactive version: theaggregate.ai/benchmark?slug=latamboard-translation-score · How the rankings work · Data refreshed daily, snapshot 2026-07-22.