OPUS-MT Dashboard

Language pair: eng-fra [all languages]
Models: [all models] [OPUS-MT] [external] [compare]
Benchmark: [all benchmarks] [average score] multi30k_test_2017_flickr
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]
include scores of user-contributed translations

Model Scores (spbleu scores on the "multi30k_test_2017_flickr" testset)

ID	spbleu	Output	Model	Size	Link
6	56.200	show	facebook/nll...illed-1.3B	1331.2M	URL
5	55.200	show	facebook/nllb-200-3.3B	3379.2M	URL
4	54.700	show	facebook/nllb-200-1.3B	1331.2M	URL
3	50.800	show	facebook/nll...illed-600M	600M	URL
2	46.900	show	facebook/m2m100_1.2B	1331.2M	URL
1	37.100	show	facebook/m2m100_418M	418M	URL
0	0.700	show	JamesH/Trans...fr_project	?	URL