This model is part of the OPUS-MT project, an effort to make neural machine translation models widely available and accessible for many languages in the world. All models are originally trained using the amazing framework of Marian NMT, an efficient NMT implementation written in pure C++. The models have been converted to pyTorch using the transformers library by huggingface. Training data is taken from OPUS and training pipelines use the procedures of OPUS-MT-train. Model Description:

Developed by: Language Technology Research Group at the University of Helsinki
Model Type: Translation (transformer-big)
Release: 2024-05-30
License: Apache-2.0
Language(s):
- Source Language(s): afr ang bar bis bzj dan deu djk drt eng enm fao frr fry gos got gsw hrx hwc icr isl jam kri ksh lim ltz nds nld nno nob non nor ofs pcm pdc pfl pih pis rop sco srm srn stq swe swg tcs tpi vls wae yid zea
- Target Language(s): deu eng fra por spa
- Valid Target Language Labels: >>deu<< >>eng<< >>fra<< >>por<< >>spa<< >>xxx<<
Original Model: opusTCv20230926max50+bt+jhubc_transformer-big_2024-05-30.zip
Resources for more information:

This is a multilingual translation model with multiple target languages. A sentence initial language token is required in the form of >>id<< (id = valid target language ID), e.g. >>deu<<

Uses

This model can be used for translation and text-to-text generation.

Risks, Limitations and Biases

CONTENT WARNING: Readers should be aware that the model is trained on various public data sets that may contain content that is disturbing, offensive, and can propagate historical and current stereotypes.

Significant research has explored bias and fairness issues with language models (see, e.g., Sheng et al. (2021) and Bender et al. (2021)).

How to Get Started With the Model

A short example code:

python

from transformers import MarianMTModel, MarianTokenizer

src_text = [
    ">>deu<< Replace this with text in an accepted source language.",
    ">>spa<< This is the second sentence."
]

model_name = "pytorch-models/opus-mt-tc-bible-big-gem-deu_eng_fra_por_spa"
tokenizer = MarianTokenizer.from_pretrained(model_name)
model = MarianMTModel.from_pretrained(model_name)
translated = model.generate(**tokenizer(src_text, return_tensors="pt", padding=True))

for t in translated:
    print( tokenizer.decode(t, skip_special_tokens=True) )

You can also use OPUS-MT models with the transformers pipelines, for example:

python

from transformers import pipeline
pipe = pipeline("translation", model="Helsinki-NLP/opus-mt-tc-bible-big-gem-deu_eng_fra_por_spa")
print(pipe(">>deu<< Replace this with text in an accepted source language."))

Training

Data: opusTCv20230926max50+bt+jhubc (source)
Pre-processing: SentencePiece (spm32k,spm32k)
Model Type: transformer-big
Original MarianNMT Model: opusTCv20230926max50+bt+jhubc_transformer-big_2024-05-30.zip
Training Scripts: GitHub Repo

Evaluation

Model scores at the OPUS-MT dashboard
test set translations: opusTCv20230926max50+bt+jhubc_transformer-big_2024-05-29.test.txt
test set scores: opusTCv20230926max50+bt+jhubc_transformer-big_2024-05-29.eval.txt
benchmark results: benchmark_results.txt
benchmark output: benchmark_translations.zip

langpair	testset	chr-F	BLEU	#sent	#words
afr-deu	tatoeba-test-v2021-08-07	0.69326	49.8	1583	9105
afr-eng	tatoeba-test-v2021-08-07	0.73601	60.6	1374	9622
afr-spa	tatoeba-test-v2021-08-07	0.73339	58.3	448	2783
dan-deu	tatoeba-test-v2021-08-07	0.73535	55.0	9998	76055
dan-eng	tatoeba-test-v2021-08-07	0.76364	63.1	10795	79684
dan-fra	tatoeba-test-v2021-08-07	0.76184	60.4	1731	11882
dan-por	tatoeba-test-v2021-08-07	0.77629	60.6	873	5360
dan-spa	tatoeba-test-v2021-08-07	0.72253	54.4	5000	35528
deu-deu	tatoeba-test-v2021-08-07	0.59994	35.5	2500	20806
deu-eng	tatoeba-test-v2021-08-07	0.66046	48.5	17565	149462
deu-fra	tatoeba-test-v2021-08-07	0.67616	49.0	12418	102721
deu-por	tatoeba-test-v2021-08-07	0.63474	42.8	10000	81482
deu-spa	tatoeba-test-v2021-08-07	0.67787	48.7	10521	82570
eng-deu	tatoeba-test-v2021-08-07	0.62438	40.8	17565	151568
eng-eng	tatoeba-test-v2021-08-07	0.80550	59.0	12062	115106
eng-fra	tatoeba-test-v2021-08-07	0.68113	49.7	12681	106378
eng-por	tatoeba-test-v2021-08-07	0.71775	52.9	13222	105265
eng-spa	tatoeba-test-v2021-08-07	0.72273	54.9	16583	134710
fao-eng	tatoeba-test-v2021-08-07	0.56661	41.2	294	1984
fry-eng	tatoeba-test-v2021-08-07	0.55624	36.5	220	1573
gos-deu	tatoeba-test-v2021-08-07	0.44090	22.9	207	1168
gos-eng	tatoeba-test-v2021-08-07	0.37900	21.1	1154	5635
gsw-eng	tatoeba-test-v2021-08-07	0.45890	31.4	205	990
isl-deu	tatoeba-test-v2021-08-07	0.67413	49.2	969	6279
isl-eng	tatoeba-test-v2021-08-07	0.66340	50.5	2503	19788
isl-spa	tatoeba-test-v2021-08-07	0.65707	46.9	238	1229
ltz-deu	tatoeba-test-v2021-08-07	0.53615	35.9	347	2208
ltz-eng	tatoeba-test-v2021-08-07	0.65361	55.5	293	1840
nds-deu	tatoeba-test-v2021-08-07	0.66127	47.6	9999	74564
nds-eng	tatoeba-test-v2021-08-07	0.62793	45.8	2500	17589
nds-fra	tatoeba-test-v2021-08-07	0.61167	43.5	857	5676
nds-por	tatoeba-test-v2021-08-07	0.59516	41.9	207	1256
nds-spa	tatoeba-test-v2021-08-07	0.64811	45.8	923	5540
nld-deu	tatoeba-test-v2021-08-07	0.72788	55.6	10218	74131
nld-eng	tatoeba-test-v2021-08-07	0.73960	60.2	12696	89978
nld-fra	tatoeba-test-v2021-08-07	0.68080	47.0	11548	82974
nld-por	tatoeba-test-v2021-08-07	0.68681	49.2	2500	17326
nld-spa	tatoeba-test-v2021-08-07	0.69773	51.4	10113	74981
nno-eng	tatoeba-test-v2021-08-07	0.68656	55.3	460	3524
nob-deu	tatoeba-test-v2021-08-07	0.69494	49.7	3525	33592
nob-eng	tatoeba-test-v2021-08-07	0.72289	58.3	4539	36823
nob-fra	tatoeba-test-v2021-08-07	0.71122	51.7	323	2269
nob-spa	tatoeba-test-v2021-08-07	0.73468	55.6	885	6866
nor-deu	tatoeba-test-v2021-08-07	0.69510	49.9	3651	34575
nor-eng	tatoeba-test-v2021-08-07	0.71970	58.0	5000	40355
nor-fra	tatoeba-test-v2021-08-07	0.72587	55.0	477	3213
nor-por	tatoeba-test-v2021-08-07	0.67981	46.3	481	4182
nor-spa	tatoeba-test-v2021-08-07	0.73337	55.7	960	7311
swe-deu	tatoeba-test-v2021-08-07	0.71563	54.8	3410	23494
swe-eng	tatoeba-test-v2021-08-07	0.75486	63.2	10362	68513
swe-fra	tatoeba-test-v2021-08-07	0.72399	57.2	1407	9580
swe-por	tatoeba-test-v2021-08-07	0.70201	49.5	320	2032
swe-spa	tatoeba-test-v2021-08-07	0.70678	54.1	1351	8235
yid-eng	tatoeba-test-v2021-08-07	0.49708	30.9	2483	15452
yid-fra	tatoeba-test-v2021-08-07	0.50306	27.5	384	2455
yid-spa	tatoeba-test-v2021-08-07	0.44196	20.8	407	2478
afr-deu	flores101-devtest	0.57773	28.6	1012	25094
afr-eng	flores101-devtest	0.74245	53.8	1012	24721
afr-fra	flores101-devtest	0.61642	35.9	1012	28343
afr-por	flores101-devtest	0.61404	35.5	1012	26519
afr-spa	flores101-devtest	0.51062	22.4	1012	29199
deu-fra	flores101-devtest	0.61951	36.0	1012	28343
deu-por	flores101-devtest	0.59947	33.2	1012	26519
deu-spa	flores101-devtest	0.52237	23.8	1012	29199
eng-deu	flores101-devtest	0.63903	36.7	1012	25094
isl-eng	flores101-devtest	0.55472	29.6	1012	24721
isl-fra	flores101-devtest	0.52253	25.7	1012	28343
isl-spa	flores101-devtest	0.45474	17.9	1012	29199
ltz-deu	flores101-devtest	0.59731	30.8	1012	25094
ltz-eng	flores101-devtest	0.63740	38.2	1012	24721
ltz-por	flores101-devtest	0.53373	26.7	1012	26519
nld-fra	flores101-devtest	0.55977	27.9	1012	28343
nld-por	flores101-devtest	0.54181	25.9	1012	26519
nld-spa	flores101-devtest	0.48625	19.6	1012	29199
swe-deu	flores101-devtest	0.59763	31.5	1012	25094
swe-eng	flores101-devtest	0.69769	46.7	1012	24721
swe-fra	flores101-devtest	0.62898	38.1	1012	28343
swe-por	flores101-devtest	0.61040	35.3	1012	26519
afr-deu	flores200-devtest	0.58114	29.2	1012	25094
afr-eng	flores200-devtest	0.74448	54.0	1012	24721
afr-fra	flores200-devtest	0.62075	37.0	1012	28343
afr-por	flores200-devtest	0.61878	36.3	1012	26519
afr-spa	flores200-devtest	0.51524	23.0	1012	29199
dan-deu	flores200-devtest	0.60226	31.4	1012	25094
dan-eng	flores200-devtest	0.70847	47.1	1012	24721
dan-fra	flores200-devtest	0.63937	38.9	1012	28343
dan-por	flores200-devtest	0.62188	36.3	1012	26519
dan-spa	flores200-devtest	0.52699	24.2	1012	29199
deu-eng	flores200-devtest	0.67277	41.6	1012	24721
deu-fra	flores200-devtest	0.62557	36.7	1012	28343
deu-por	flores200-devtest	0.60450	34.1	1012	26519
deu-spa	flores200-devtest	0.52346	24.0	1012	29199
eng-deu	flores200-devtest	0.64294	37.3	1012	25094
eng-fra	flores200-devtest	0.70649	49.0	1012	28343
eng-por	flores200-devtest	0.70802	48.9	1012	26519
eng-spa	flores200-devtest	0.55605	27.1	1012	29199
fao-deu	flores200-devtest	0.42553	16.2	1012	25094
fao-eng	flores200-devtest	0.45067	22.5	1012	24721
fao-fra	flores200-devtest	0.40660	17.5	1012	28343
fao-por	flores200-devtest	0.40555	16.5	1012	26519
isl-deu	flores200-devtest	0.49977	21.9	1012	25094
isl-eng	flores200-devtest	0.56358	30.4	1012	24721
isl-fra	flores200-devtest	0.53181	26.6	1012	28343
isl-por	flores200-devtest	0.51706	25.3	1012	26519
isl-spa	flores200-devtest	0.46015	18.4	1012	29199
lim-deu	flores200-devtest	0.44990	15.8	1012	25094
lim-eng	flores200-devtest	0.47463	21.5	1012	24721
lim-fra	flores200-devtest	0.44097	18.2	1012	28343
lim-por	flores200-devtest	0.43247	17.2	1012	26519
ltz-deu	flores200-devtest	0.60196	31.3	1012	25094
ltz-eng	flores200-devtest	0.64373	38.9	1012	24721
ltz-fra	flores200-devtest	0.59216	33.4	1012	28343
ltz-por	flores200-devtest	0.53423	26.8	1012	26519
ltz-spa	flores200-devtest	0.45465	18.2	1012	29199
nld-deu	flores200-devtest	0.53629	22.3	1012	25094
nld-eng	flores200-devtest	0.59199	30.7	1012	24721
nld-fra	flores200-devtest	0.56488	28.4	1012	28343
nld-por	flores200-devtest	0.54554	26.2	1012	26519
nld-spa	flores200-devtest	0.48955	20.0	1012	29199
nno-deu	flores200-devtest	0.57440	28.4	1012	25094
nno-eng	flores200-devtest	0.68265	43.8	1012	24721
nno-fra	flores200-devtest	0.60865	35.6	1012	28343
nno-por	flores200-devtest	0.59792	33.8	1012	26519
nno-spa	flores200-devtest	0.51118	22.8	1012	29199
nob-deu	flores200-devtest	0.56226	26.6	1012	25094
nob-eng	flores200-devtest	0.67302	42.9	1012	24721
nob-fra	flores200-devtest	0.60587	34.3	1012	28343
nob-por	flores200-devtest	0.59086	32.4	1012	26519
nob-spa	flores200-devtest	0.50910	22.0	1012	29199
swe-deu	flores200-devtest	0.59867	31.7	1012	25094
swe-eng	flores200-devtest	0.70155	47.4	1012	24721
swe-fra	flores200-devtest	0.63360	39.0	1012	28343
swe-por	flores200-devtest	0.61265	35.8	1012	26519
swe-spa	flores200-devtest	0.52209	24.1	1012	29199
tpi-deu	flores200-devtest	0.40345	10.6	1012	25094
tpi-eng	flores200-devtest	0.48317	19.5	1012	24721
tpi-fra	flores200-devtest	0.43681	16.1	1012	28343
tpi-por	flores200-devtest	0.43402	16.0	1012	26519
deu-eng	generaltest2022	0.55482	30.2	1984	37634
deu-fra	generaltest2022	0.60682	37.2	1984	38276
eng-deu	generaltest2022	0.60337	32.5	2037	38914
deu-eng	multi30k_test_2016_flickr	0.60928	40.0	1000	12955
deu-fra	multi30k_test_2016_flickr	0.62274	37.9	1000	13505
eng-deu	multi30k_test_2016_flickr	0.63626	34.7	1000	12106
eng-fra	multi30k_test_2016_flickr	0.69972	48.3	1000	13505
deu-eng	multi30k_test_2017_flickr	0.63346	40.7	1000	11374
deu-fra	multi30k_test_2017_flickr	0.62340	36.9	1000	12118
eng-deu	multi30k_test_2017_flickr	0.62213	33.2	1000	10755
eng-fra	multi30k_test_2017_flickr	0.71289	50.2	1000	12118
deu-eng	multi30k_test_2017_mscoco	0.55818	32.5	461	5231
deu-fra	multi30k_test_2017_mscoco	0.59452	34.4	461	5484
eng-deu	multi30k_test_2017_mscoco	0.57860	29.2	461	5158
eng-fra	multi30k_test_2017_mscoco	0.72971	54.1	461	5484
deu-eng	multi30k_test_2018_flickr	0.59041	36.1	1071	14689
deu-fra	multi30k_test_2018_flickr	0.56956	30.6	1071	15867
eng-deu	multi30k_test_2018_flickr	0.59517	30.5	1071	13703
eng-fra	multi30k_test_2018_flickr	0.64575	40.6	1071	15867
eng-fra	newsdiscusstest2015	0.63295	37.4	1500	27975
deu-eng	newssyscomb2009	0.55577	29.0	502	11818
deu-fra	newssyscomb2009	0.55843	27.4	502	12331
deu-spa	newssyscomb2009	0.55364	28.0	502	12503
eng-deu	newssyscomb2009	0.53472	22.7	502	11271
eng-fra	newssyscomb2009	0.58209	29.0	502	12331
eng-spa	newssyscomb2009	0.58210	31.1	502	12503
deu-eng	newstest2008	0.54406	26.9	2051	49380
deu-fra	newstest2008	0.54711	25.8	2051	52685
deu-spa	newstest2008	0.53767	25.4	2051	52586
eng-deu	newstest2008	0.52827	23.0	2051	47447
eng-fra	newstest2008	0.55113	26.4	2051	52685
eng-spa	newstest2008	0.56379	29.5	2051	52586
deu-eng	newstest2009	0.53786	26.3	2525	65399
deu-fra	newstest2009	0.53993	25.6	2525	69263
deu-spa	newstest2009	0.53437	25.3	2525	68111
eng-deu	newstest2009	0.52924	22.0	2525	62816
eng-fra	newstest2009	0.57445	28.9	2525	69263
eng-spa	newstest2009	0.57630	29.7	2525	68111
deu-eng	newstest2010	0.58260	30.2	2489	61711
deu-fra	newstest2010	0.57553	28.6	2489	66022
deu-spa	newstest2010	0.59093	32.1	2489	65480
eng-deu	newstest2010	0.54601	25.2	2489	61503
eng-fra	newstest2010	0.59310	31.5	2489	66022
eng-spa	newstest2010	0.61660	36.1	2489	65480
deu-eng	newstest2011	0.54865	26.6	3003	74681
deu-fra	newstest2011	0.55548	27.0	3003	80626
deu-spa	newstest2011	0.56405	30.1	3003	79476
eng-deu	newstest2011	0.52308	22.5	3003	72981
eng-fra	newstest2011	0.60656	33.6	3003	80626
eng-spa	newstest2011	0.61760	38.1	3003	79476
deu-eng	newstest2012	0.56119	28.1	3003	72812
deu-fra	newstest2012	0.55823	27.1	3003	78011
deu-spa	newstest2012	0.57080	30.9	3003	79006
eng-deu	newstest2012	0.52461	23.2	3003	72886
eng-fra	newstest2012	0.58890	31.2	3003	78011
eng-spa	newstest2012	0.62050	38.2	3003	79006
deu-eng	newstest2013	0.57911	31.4	3000	64505
deu-fra	newstest2013	0.56212	29.6	3000	70037
deu-spa	newstest2013	0.56814	31.4	3000	70528
eng-deu	newstest2013	0.55370	27.0	3000	63737
eng-fra	newstest2013	0.58712	33.0	3000	70037
eng-spa	newstest2013	0.59653	35.0	3000	70528
deu-eng	newstest2014	0.59441	32.7	3003	67337
eng-deu	newstest2014	0.58610	27.3	3003	62688
eng-fra	newstest2014	0.64743	38.9	3003	77306
deu-eng	newstest2015	0.59573	33.6	2169	46443
eng-deu	newstest2015	0.59581	31.3	2169	44260
deu-eng	newstest2016	0.64594	40.0	2999	64119
eng-deu	newstest2016	0.63617	36.8	2999	62669
deu-eng	newstest2017	0.60931	35.5	3004	64399
eng-deu	newstest2017	0.58459	29.3	3004	61287
deu-eng	newstest2018	0.66399	42.9	2998	67012
eng-deu	newstest2018	0.68676	44.7	2998	64276
deu-eng	newstest2019	0.63254	39.0	2000	39227
deu-fra	newstest2019	0.62658	35.7	1701	42509
eng-deu	newstest2019	0.65659	40.9	1997	48746
deu-eng	newstest2020	0.60708	34.3	785	38220
deu-fra	newstest2020	0.60215	32.7	1619	36890
eng-deu	newstest2020	0.59764	31.2	1418	52383
deu-eng	newstest2021	0.59635	31.7	1000	20180
deu-fra	newstest2021	0.58868	31.0	1000	23757
eng-deu	newstest2021	0.56042	25.5	1002	27970
isl-eng	newstest2021	0.54878	30.5	1000	22529
deu-eng	newstestALL2020	0.60708	34.3	785	38220
eng-deu	newstestALL2020	0.59764	31.2	1418	52383
deu-eng	newstestB2020	0.60692	34.1	785	37696
eng-deu	newstestB2020	0.58796	30.6	1418	53092
afr-deu	ntrex128	0.55455	26.1	1997	48761
afr-eng	ntrex128	0.72478	51.5	1997	47673
afr-fra	ntrex128	0.56551	29.1	1997	53481
afr-por	ntrex128	0.55192	28.0	1997	51631
afr-spa	ntrex128	0.58404	33.5	1997	54107
dan-deu	ntrex128	0.53830	24.6	1997	48761
dan-eng	ntrex128	0.62590	38.0	1997	47673
dan-fra	ntrex128	0.53853	26.2	1997	53481
dan-por	ntrex128	0.53295	26.6	1997	51631
dan-spa	ntrex128	0.55771	30.2	1997	54107
deu-eng	ntrex128	0.61495	33.5	1997	47673
deu-fra	ntrex128	0.55631	28.1	1997	53481
deu-por	ntrex128	0.54078	25.6	1997	51631
deu-spa	ntrex128	0.56793	30.6	1997	54107
eng-deu	ntrex128	0.58373	29.7	1997	48761
eng-fra	ntrex128	0.61168	34.9	1997	53481
eng-por	ntrex128	0.59703	33.3	1997	51631
eng-spa	ntrex128	0.63737	39.8	1997	54107
fao-deu	ntrex128	0.42430	16.0	1997	48761
fao-eng	ntrex128	0.47552	24.2	1997	47673
fao-fra	ntrex128	0.40501	16.8	1997	53481
fao-por	ntrex128	0.40261	15.9	1997	51631
fao-spa	ntrex128	0.41996	18.2	1997	54107
isl-deu	ntrex128	0.49197	19.9	1997	48761
isl-eng	ntrex128	0.55641	28.4	1997	47673
isl-fra	ntrex128	0.50981	23.2	1997	53481
isl-por	ntrex128	0.49008	20.9	1997	51631
isl-spa	ntrex128	0.52176	26.1	1997	54107
ltz-deu	ntrex128	0.51598	22.0	1997	48761
ltz-eng	ntrex128	0.58852	33.1	1997	47673
ltz-fra	ntrex128	0.50270	22.5	1997	53481
ltz-por	ntrex128	0.46568	19.7	1997	51631
ltz-spa	ntrex128	0.48287	22.3	1997	54107
nld-deu	ntrex128	0.55925	25.7	1997	48761
nld-eng	ntrex128	0.63556	36.2	1997	47673
nld-fra	ntrex128	0.55930	27.7	1997	53481
nld-por	ntrex128	0.54481	27.1	1997	51631
nld-spa	ntrex128	0.57597	32.0	1997	54107
nno-deu	ntrex128	0.52772	24.1	1997	48761
nno-eng	ntrex128	0.65262	41.9	1997	47673
nno-fra	ntrex128	0.54074	27.0	1997	53481
nno-por	ntrex128	0.52446	25.3	1997	51631
nno-spa	ntrex128	0.55555	30.1	1997	54107
nob-deu	ntrex128	0.55215	26.2	1997	48761
nob-eng	ntrex128	0.64785	39.6	1997	47673
nob-fra	ntrex128	0.55925	28.9	1997	53481
nob-por	ntrex128	0.54040	26.1	1997	51631
nob-spa	ntrex128	0.57388	31.9	1997	54107
swe-deu	ntrex128	0.55195	26.2	1997	48761
swe-eng	ntrex128	0.65697	41.4	1997	47673
swe-fra	ntrex128	0.56103	29.1	1997	53481
swe-por	ntrex128	0.54574	27.3	1997	51631
swe-spa	ntrex128	0.57745	32.3	1997	54107
eng-fra	tico19-test	0.62405	38.9	2100	64661
eng-por	tico19-test	0.72533	49.5	2100	62729
eng-spa	tico19-test	0.72498	51.0	2100	66563

Citation Information

Publications: Democratizing neural machine translation with OPUS-MT and OPUS-MT – Building open translation services for the World and The Tatoeba Translation Challenge – Realistic Data Sets for Low Resource and Multilingual MT (Please, cite if you use this model.)

bibtex

@article{tiedemann2023democratizing,
  title={Democratizing neural machine translation with {OPUS-MT}},
  author={Tiedemann, J{\"o}rg and Aulamo, Mikko and Bakshandaeva, Daria and Boggia, Michele and Gr{\"o}nroos, Stig-Arne and Nieminen, Tommi and Raganato, Alessandro and Scherrer, Yves and Vazquez, Raul and Virpioja, Sami},
  journal={Language Resources and Evaluation},
  number={58},
  pages={713--755},
  year={2023},
  publisher={Springer Nature},
  issn={1574-0218},
  doi={10.1007/s10579-023-09704-w}
}

@inproceedings{tiedemann-thottingal-2020-opus,
    title = "{OPUS}-{MT} {--} Building open translation services for the World",
    author = {Tiedemann, J{\"o}rg  and Thottingal, Santhosh},
    booktitle = "Proceedings of the 22nd Annual Conference of the European Association for Machine Translation",
    month = nov,
    year = "2020",
    address = "Lisboa, Portugal",
    publisher = "European Association for Machine Translation",
    url = "https://aclanthology.org/2020.eamt-1.61",
    pages = "479--480",
}

@inproceedings{tiedemann-2020-tatoeba,
    title = "The Tatoeba Translation Challenge {--} Realistic Data Sets for Low Resource and Multilingual {MT}",
    author = {Tiedemann, J{\"o}rg},
    booktitle = "Proceedings of the Fifth Conference on Machine Translation",
    month = nov,
    year = "2020",
    address = "Online",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2020.wmt-1.139",
    pages = "1174--1182",
}

Acknowledgements

The work is supported by the HPLT project, funded by the European Union’s Horizon Europe research and innovation programme under grant agreement No 101070350. We are also grateful for the generous computational resources and IT infrastructure provided by CSC -- IT Center for Science, Finland, and the EuroHPC supercomputer LUMI.

Model conversion info

transformers version: 4.45.1
OPUS-MT git hash: 0882077
port time: Tue Oct 8 11:04:22 EEST 2024
port machine: LM0-400-22516.local

WindstormLabs/origin-Helsinki-NLP-opus-mt-tc-bible-big-gem-deu_eng_fra_por_spa

opus-mt-tc-bible-big-gem-deu_eng_fra_por_spa

Table of Contents

Model Details