OPUS-MT Dashboard

Language pair: deu-eng [all languages]
Models: [all models] [OPUS-MT] [external] [compare]
Benchmark: [all benchmarks] [average score] multi30k_test_2017_mscoco
Evaluation metric: bleu [spbleu][chrf][chrf++][comet]
Chart Type: [barchart][scatterplot] [size] [type]

orange = OPUS-MT, blue = Tatoeba-MT models, red = HPLT-MT models
green = student models, grey = external models, purple = user-contributed
render chart with [gd] [plotly]

Model Scores (bleu scores on the "multi30k_test_2017_mscoco" testset)

ID	bleu	Output	Model	Size	Link
8	37.0	show	facebook/nllb-200-3.3B	3379.2M	URL
7	36.6	show	facebook/nllb-200-1.3B	1331.2M	URL
6	36.4	show	facebook/wmt19-de-en	?	URL
5	36.4	show	facebook/nll...illed-1.3B	1331.2M	URL
4	34.2	show	allenai/wmt1...en-6-6-big	?	URL
3	33.7	show	facebook/nll...illed-600M	600M	URL
2	29.8	show	allenai/wmt1...n-6-6-base	?	URL
1	29.7	show	facebook/m2m100_1.2B	1331.2M	URL
0	23.5	show	facebook/m2m100_418M	418M	URL