How to use from the
Use from the
Transformers.js library
// npm i @huggingface/transformers
import { pipeline } from '@huggingface/transformers';

// Allocate pipeline
const pipe = await pipeline('translation', 'OptimaMT/es-ru');

OptimaMT es β†’ ru (Spanish to Russian)

Drop-in copy of Xenova/opus-mt-es-ru at revision 268ccd60bde3 with a lighter merged decoder: ONNX Runtime's fast mode (disable_quant_qdq = 1, 2.3–2.5Γ— faster) at half the memory. Same weights, same tokens. Part of OptimaMT: all 51 pairs, named OptimaMT/<src>-<tgt>.

Used in production:

  • OpenCircle, a desktop screen translator for Windows and Linux, ships these models since v0.1.4.
Memory after load Linux Windows Identical tokens
disable_quant_qdq = 1 532 β†’ 284 MiB (-47 %) 508 β†’ 259 MiB (-49 %) yes
ONNX Runtime defaults 284 β†’ 222 MiB (-22 %) 201 β†’ 166 MiB (-17 %) yes

FLORES-200 devtest (1,012 sentences): the same translations as the original, BLEU 14.77, chrF 44.71 (greedy).

ONNX Runtime 1.31.1, CPU, 4 threads (Linux; Windows in a VM). Interactive results: OptimaMT/results Β· method: REPORT.md.

Use

import { pipeline } from "@huggingface/transformers";

const translator = await pipeline("translation", "OptimaMT/es-ru", {
  dtype: "q8",
  session_options: { extra: { session: { disable_quant_qdq: "1" } } },
});
const [out] = await translator("El faro estΓ‘ en la costa.");
console.log(out.translation_text);

Python (translate.py: onnxruntime, sentencepiece, numpy):

python translate.py OptimaMT/es-ru "El faro estΓ‘ en la costa."

Files

Only onnx/decoder_model_merged_quantized.onnx is rebuilt; every other file is the original, byte for byte. Vocabulary: 63,430 tokens.

Licence and credits

  • Weights: OPUS-MT by the University of Helsinki, CC BY 4.0.
  • ONNX export: Xenova/opus-mt-es-ru at revision 268ccd60bde3.
  • Modified: the decoder's output projection was moved out of its two branches (graph change only). CC BY 4.0.
Downloads last month
107
Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support

Model tree for OptimaMT/es-ru

Quantized
(1)
this model

Space using OptimaMT/es-ru 1

Collection including OptimaMT/es-ru