Downloads · 30 days
9
10% of all-time downloads
NexaAI/HY-MT1.5-1.8B-npu
HY-MT1.5-1.8B-npu is a machine learning model from NexaAI. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
HY-MT1.5-1.8B is a 1.8B-parameter multilingual neural machine translation model released by Tencent as part of the Hunyuan Translation Model v1.5 series.
Downloads · 30 days
9
10% of all-time downloads
All-time downloads
86
Public
Repo size
2.8 GB
Likes
1
Public
Click a slice to open those files.
.nexa2.8 GB · 100%
From the Hugging Face model README
HY-MT1.5-1.8B is a 1.8B-parameter multilingual neural machine translation model released by Tencent as part of the Hunyuan Translation Model v1.5 series.
The model focuses on high-quality, high-speed translation across 33 major languages plus 5 ethnic/dialect variants, achieving performance comparable to the larger 7B model while remaining lightweight. After quantization, HY-MT1.5-1.8B is suitable for edge and on-device deployment, enabling real-time translation scenarios.
nexa config set license '<access_token>'
nexa infer NexaAI/LFM2.5-1.2B-npu
Multilingual translation: supports 36 languages, including Chinese, English, European, Middle Eastern, South Asian, and Southeast Asian languages.
Small yet strong: industry-leading translation quality among ~2B-parameter models, surpassing many commercial translation APIs.
Edge-ready: supports FP8 and INT4 quantization for efficient deployment on resource-constrained devices.
Advanced translation controls:
Robust scenarios: strong performance on mixed-language and explanatory translation tasks.
This repo is licensed under the Creative Commons Attribution–NonCommercial 4.0 (CC BY-NC 4.0) license, which allows use, sharing, and modification only for non-commercial purposes with proper attribution. All NPU-related models, runtimes, and code in this project are protected under this non-commercial license and cannot be used in any commercial or revenue-generating applications. Commercial licensing or enterprise usage requires a separate agreement. For inquiries, please contact [email protected]