software-mansion/react-native-executorch-opus-mt

opus-mt

This repository hosts the opus-mt models exported for the React Native ExecuTorch library as ExecuTorch .pte programs, ready to run on device.

Variants

Path	Backend	Precision
en_de/xnnpack/opus_mt_en_de_xnnpack_int8.pte	xnnpack	int8
de_en/xnnpack/opus_mt_de_en_xnnpack_int8.pte	xnnpack	int8
en_es/xnnpack/opus_mt_en_es_xnnpack_int8.pte	xnnpack	int8
es_en/xnnpack/opus_mt_es_en_xnnpack_int8.pte	xnnpack	int8
en_fr/xnnpack/opus_mt_en_fr_xnnpack_int8.pte	xnnpack	int8
fr_en/xnnpack/opus_mt_fr_en_xnnpack_int8.pte	xnnpack	int8

Unshipped backends

A backend directory carrying a NOTES.md and no .pte was exported and not published. Its note records why.

• de_en/coreml (why)
• de_en/mlx (why)
• de_en/vulkan (why)
• en_de/coreml (why)
• en_de/mlx (why)
• en_de/vulkan (why)
• en_es/coreml (why)
• en_es/mlx (why)
• en_es/vulkan (why)
• en_fr/coreml (why)
• en_fr/mlx (why)
• en_fr/vulkan (why)
• es_en/coreml (why)
• es_en/mlx (why)
• es_en/vulkan (why)
• fr_en/coreml (why)
• fr_en/mlx (why)
• fr_en/vulkan (why)

Repository structure

LICENSE                                        360 B
config.json                                    29 B
de_en/coreml/NOTES.md                          230 B
de_en/mlx/NOTES.md                             169 B
de_en/tokenizer.json                           2.4 MB
de_en/vulkan/NOTES.md                          218 B
de_en/xnnpack/config.json                      5.5 kB
de_en/xnnpack/opus_mt_de_en_xnnpack_int8.pte   107.0 MB
en_de/coreml/NOTES.md                          230 B
en_de/mlx/NOTES.md                             169 B
en_de/tokenizer.json                           2.4 MB
en_de/vulkan/NOTES.md                          218 B
en_de/xnnpack/config.json                      5.5 kB
en_de/xnnpack/opus_mt_en_de_xnnpack_int8.pte   107.0 MB
en_es/coreml/NOTES.md                          230 B
en_es/mlx/NOTES.md                             169 B
en_es/tokenizer.json                           2.7 MB
en_es/vulkan/NOTES.md                          218 B
en_es/xnnpack/config.json                      5.5 kB
en_es/xnnpack/opus_mt_en_es_xnnpack_int8.pte   114.1 MB
en_fr/coreml/NOTES.md                          230 B
en_fr/mlx/NOTES.md                             169 B
en_fr/tokenizer.json                           2.4 MB
en_fr/vulkan/NOTES.md                          218 B
en_fr/xnnpack/config.json                      5.5 kB
en_fr/xnnpack/opus_mt_en_fr_xnnpack_int8.pte   108.4 MB
es_en/coreml/NOTES.md                          230 B
es_en/mlx/NOTES.md                             169 B
es_en/tokenizer.json                           2.7 MB
es_en/vulkan/NOTES.md                          218 B
es_en/xnnpack/config.json                      5.5 kB
es_en/xnnpack/opus_mt_es_en_xnnpack_int8.pte   114.1 MB
fr_en/coreml/NOTES.md                          230 B
fr_en/mlx/NOTES.md                             169 B
fr_en/tokenizer.json                           2.4 MB
fr_en/vulkan/NOTES.md                          218 B
fr_en/xnnpack/config.json                      5.5 kB
fr_en/xnnpack/opus_mt_fr_en_xnnpack_int8.pte   108.4 MB

Compatibility

These files are published for the ExecuTorch v1.4.1 runtime. ExecuTorch gives no forward compatibility guarantee, so an older runtime may fail to load them.

To use them in React Native ExecuTorch, pass the model constant shipped in the library's model registry to the corresponding task pipeline. See the documentation.

To load these files in your own ExecuTorch runtime, read the compatibility note first.

Upstream models and licenses

Each directory is one language pair with its own upstream model and license. en_de is CC-BY 4.0 and requires attribution to Helsinki-NLP Opus-MT.

Directory	Upstream	License
en_de	Helsinki-NLP/opus-mt-en-de	cc-by-4.0
de_en	Helsinki-NLP/opus-mt-de-en	apache-2.0
en_es	Helsinki-NLP/opus-mt-en-es	apache-2.0
es_en	Helsinki-NLP/opus-mt-es-en	apache-2.0
en_fr	Helsinki-NLP/opus-mt-en-fr	apache-2.0
fr_en	Helsinki-NLP/opus-mt-fr-en	apache-2.0

Export notes

Each .pte has two methods: encode returns the cross-attention keys and values of all decoder layers, and decode runs one greedy step against a 256-token self-attention cache held in the program. The tokenizer adds no </s>; append it, start decoding from <pad> and stop at </s>. Only XNNPACK int8 ships: it was the fastest variant on both a Galaxy S26 Ultra and an iPhone 16.