Instructions to use vlad-grig/opus-mt-en-it with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers.js
How to use vlad-grig/opus-mt-en-it with Transformers.js:
// npm i @huggingface/transformers import { pipeline } from '@huggingface/transformers'; // Allocate pipeline const pipe = await pipeline('translation', 'vlad-grig/opus-mt-en-it');
opus-mt-en-it, ONNX weights for the browser
English to Italian. ONNX weights so the model runs in a web browser through Transformers.js.
This repository exists so that a WordPress plugin that translates product catalogues in the shop owner's own browser does not depend on a repository it does not control. Nothing here is new work.
Where it comes from
- The model is Helsinki-NLP/opus-mt-en-it, trained by the Language Technology Research Group at the University of Helsinki, released under CC-BY-4.0.
- The ONNX conversion follows Xenova/opus-mt-en-it.
What is here
Only the weights a browser actually loads: fp32 for WebGPU and q8 for the
WebAssembly fallback. The other quantisations are left out on purpose, because
nothing downloads them and they would cost a visitor's bandwidth to no end.
Licence
CC-BY-4.0, as the original. Attribution above.
- Downloads last month
- 40
Model tree for vlad-grig/opus-mt-en-it
Base model
Helsinki-NLP/opus-mt-en-it