Seed-X: Building Strong Multilingual Translation LLM with 7B Parameters

TL;DR

Seed-X introduces a 7B parameter multilingual translation LLM that leverages diverse data, Chain-of-Thought reasoning, and reinforcement learning to achieve state-of-the-art performance across 28 languages, rivaling larger closed-source models.

Contribution

The paper presents Seed-X, a novel open-source multilingual translation LLM with innovative training and fine-tuning strategies for improved translation quality.

Findings

Seed-X matches performance of Gemini-2.5 and GPT-4o.

Outperforms larger open-source models in metrics and human evaluations.

Effective use of Chain-of-Thought reasoning and reinforcement learning.

Abstract

Multilingual translation stands as a challenging task for large language models (LLMs) to handle intricate language patterns and stilted translations that arise in automated translations. In this paper, we introduce Seed-X, a family of open-source LLMs comprising instruct and reasoning models, pushing the limits of translation capability with 7B parameter size. The base model is pre-trained on a diverse, high-quality dataset encompassing both monolingual and bilingual content across 28 languages, harnessing the full potential of multilingual data. The instruct model is then finetuned to translate by Chain-of-Thought (CoT) reasoning and further enhanced through reinforcement learning (RL) to achieve better generalization across diverse language pairs. Seed-X achieves performance comparable to leading closed-source models, including Gemini-2.5 and GPT-4o, across 28 languages, and significantly outperforms larger open-source models in both automatic metrics and human evaluations. We share the best practices through our optimization process, and make the parameter public available for advancing translation research and applications.