OPUS-MT Dashboard

Language pair: fra-ces [all languages]
Models: [all models] [OPUS-MT] [external] [contributed] [compare]
Benchmark: [all benchmarks] [average score] tatoeba-test-v2020-07-28
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]
exclude scores of user-contributed translations

Model Scores (spbleu scores on the "tatoeba-test-v2020-07-28" testset)

ID	spbleu	Output	Model	Size	Link
10	55.100	show	deu+eng+fra+...2024-05-30	237.8M	zip-file
9	54.600	show	fra-ces/opus-2021-02-18	74.5M	zip-file
8	52.200	show	facebook/nll...illed-1.3B	1331.2M	URL
7	50.800	show	fra-sla/opus...2022-09-15	238.8M	zip-file
6	50.200	show	facebook/nllb-200-3.3B	3379.2M	URL
5	49.500	show	deu+eng+fra+...2024-05-30	231M	zip-file
4	45.700	show	facebook/nll...illed-600M	600M	URL
3	43.000	show	deu+eng+fra+...2024-05-30	240.9M	zip-file
2	42.900	show	facebook/m2m100_418M	418M	URL
1	39.400	show	facebook/nllb-200-1.3B	1331.2M	URL
0	38.600	show	mul-mul/opus...2024-08-17	247.8M	zip-file