OptimaMT en β†’ es (English to Spanish)

Drop-in copy of Xenova/opus-mt-en-es at revision 4b002a4c7edd 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 = 1 545 β†’ 291 MiB (-47 %) 517 β†’ 267 MiB (-48 %) yes
ONNX Runtime defaults 291 β†’ 227 MiB (-22 %) 201 β†’ 170 MiB (-15 %) yes

FLORES-200 devtest (1,012 sentences): the same translations as the original, BLEU 26.15, chrF 54.81 (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/en-es", {
  dtype: "q8",
  session_options: { extra: { session: { disable_quant_qdq: "1" } } },
});
const [out] = await translator("The lighthouse stands on the coast.");
console.log(out.translation_text);

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

python translate.py OptimaMT/en-es "The lighthouse stands on the coast."

Files

Only onnx/decoder_model_merged_quantized.onnx is rebuilt; every other file is the original, byte for byte. Vocabulary: 65,001 tokens.

Licence and credits

  • Weights: OPUS-MT by the University of Helsinki, CC BY 4.0.
  • ONNX export: Xenova/opus-mt-en-es at revision 4b002a4c7edd.
  • Modified: the decoder's output projection was moved out of its two branches (graph change only). CC BY 4.0.
Downloads last month
113
Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support

Model tree for OptimaMT/en-es

Quantized
(2)
this model

Space using OptimaMT/en-es 1

Collection including OptimaMT/en-es