opus-mt-fr-he β€” ONNX (int8) for transformers.js

ONNX export (8-bit dynamic quantization) of Helsinki-NLP/opus-mt-fr-he (Tiedemann et al., OPUS-MT, Apache-2.0), with a tokenizer.json so it runs in the browser with transformers.js:

const t = await pipeline("translation", "sivelsguy/opus-mt-fr-he-onnx", { dtype: "q8" });

Translate one sentence per call: batched inputs of different lengths may end with repeated punctuation.

Used by the open-source Hebrew learning app Parle. All credit for the model goes to Helsinki-NLP.

Downloads last month
33
Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support

Model tree for sivelsguy/opus-mt-fr-he-onnx

Quantized
(1)
this model