做多语言翻译或端侧部署的团队终于有了一个又快又准的选择——1.8B 量化后 440MB 就能跑，性能还超过微软豆包 API，值得直接上手试。
Hy-MT2 是一系列面向复杂真实场景的快速多语言翻译模型，包含 1.8B、7B 和 30B-A3B（MoE）三个尺寸，支持 33 种语言间的翻译。通过 AngelSlim 1.25-bit 极端量化，1.8B 模型仅需 440MB 存储，推理速度提升 1.5 倍，适合端侧部署。在通用、商业、领域和指令跟随翻译任务中，7B 和 30B 模型在快速思考模式下超越 DeepSeek-V4-Pro 和 Kimi K2.6 等开源模型，1.8B 模型整体也优于微软和豆包等主流商业 API。该工作为多语言翻译提供了高效且强大的新选择。
Hy-MT2: A Family of Fast, Efficient and Powerful Multilingual Translation Models in the Wild
Hy-MT2 is a family of fast-thinking multilingual translation models designed for complex real-world scenarios. It includes three model sizes: 1.8B, 7B, and 30B-A3B (MoE), all of which support translation among 33 languages and effectively follow translation instructions in multiple languages. For on-device deployment, with AngelSlim 1.25-bit extreme quantization, the 1.8B model requires only 440 MB of storage and improves inference speed by 1.5x. Multi-dimensional evaluations show that Hy-MT2 delivers outstanding performance across general, real-world business, domain-specific, and instruction-following translation tasks. The 7B and 30B models outperform open-source models such as DeepSeek-V4-Pro and Kimi K2.6 in fast-thinking mode, while the lightweight 1.8B model also surpasses mainstream commercial APIs from providers such as Microsoft and Doubao overall.
- 岚叔05-22 11:05原文