OPUS-MT Dashboard

blue = OPUS-MT / Tatoeba-MT models, grey = external models, purple = user-contributed
render chart with [gd] [plotly]

ID	Benchmark (bleu)	Output	OPUS-MT	bleu	external	bleu	Diff
0	flores101-devtest	compare	eng-slv/opus...2021-05-15	29.2	facebook/m2m100_1.2B	28.2	1.0
1	flores200-devtest	compare	deu+eng+fra+...2024-05-30	31.3	facebook/m2m100_1.2B	28.2	3.1
2	ntrex128	compare	deu+eng+fra+...2024-05-30	30.2	facebook/nll...illed-1.3B	25.9	4.3
3	tatoeba-test-v2020-07-28	compare	en-sl/opus-2019-12-04	20.2	facebook/nllb-200-3.3B	19	1.2
4	tatoeba-test-v2021-03-30	compare	en-sl/opus-2019-12-04	20.2	facebook/nllb-200-3.3B	18.9	1.3
5	tatoeba-test-v2021-08-07	compare	en-sl/opus-2019-12-04	17.1	facebook/nll...illed-1.3B	15.8	1.3
		average		24.7		22.7	2.0