# Multilingual Evaluation

A large-scale language model specialized for translation tasks developed by Preferred Networks, supporting English and Japanese mutual translation

M-Prometheus is an open-source LLM evaluation model capable of natively supporting multilingual output assessment. Trained on 480,000 multilingual direct evaluation and pairwise comparison data with long-text feedback.

M-Prometheus is an open-source LLM evaluation model capable of natively supporting multilingual output evaluation. It was trained on 480,000 multilingual direct evaluation and paired comparison data points.

M-Prometheus is an open-source LLM evaluation model that natively supports multilingual output evaluation. It is trained on 480,000 multilingual direct evaluation and pairwise comparison data points with detailed feedback.

Atla Selene Mini is currently the most advanced small judge language model (SLMJ), with performance comparable to models 10 times its size, surpassing GPT-4o in multiple benchmarks.

UniTE is a unified framework for translation quality evaluation, supporting multilingual translation assessment.