@AdinaYakup: Hy-MT2 New translation model family from @TencentHunyuan 1.8B / 7B / 30B-A3B MoE Supports 33 languages 1.8B > 440MB wit…

X AI KOLs Following Models

Summary

Tencent Hunyuan released Hy-MT2, a family of translation models up to 30B parameters with MoE, supporting 33 languages and quantized for on-device use.

Hy-MT2 🔥 New translation model family from @TencentHunyuan ✨ 1.8B / 7B / 30B-A3B MoE ✨ Supports 33 languages ✨ 1.8B > 440MB with 1.25-bit quantization ✨ Runs on device with faster inference ✨ 1.8B outperforms some commercial APIs https://t.co/Ep7gL8wedk
Original Article
View Cached Full Text

Cached at: 05/22/26, 03:53 AM

Hy-MT2 🔥 New translation model family from @TencentHunyuan

✨ 1.8B / 7B / 30B-A3B MoE ✨ Supports 33 languages ✨ 1.8B > 440MB with 1.25-bit quantization ✨ Runs on device with faster inference ✨ 1.8B outperforms some commercial APIs https://t.co/Ep7gL8wedk

Similar Articles

BuzzASR: A Swarm of 100+ Monolingual Speech Recognition Models

arXiv cs.CL

BuzzASR is a collection of language-specialized Whisper models for automatic speech recognition in 102 languages, outperforming Whisper-large-v3 on 77 languages with significant improvements in error rates and compression efficiency.