OPUS-MT Dashboard

Language pair: eng-ces
Models: [all models] [OPUS-MT] [external] [compare]
Benchmark: all benchmarks [average score]
Evaluation metric: bleu [spbleu][chrf][chrf++][comet]
Chart Type: [barchart][diff]

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

Model Scores (comparing between OPUS-MT and external models)

ID	Benchmark (bleu)	Output	OPUS-MT	bleu	external	bleu	Diff
0	flores101-devtest	compare	eng-ces+slk/...2022-03-13	34.1	facebook/m2m100_1.2B	30.4	3.7
1	flores200-devtest	compare	eng-ces+slk/...2022-03-13	34.1	facebook/m2m100_1.2B	30.4	3.7
2	generaltest2022	compare	eng-ces+slk/...2022-03-13	44.9	facebook/nllb-200-3.3B	39.4	5.5
3	multi30k_test_2016_flickr	compare	eng-ces+slk/...2022-03-13	33.4	facebook/nll...illed-1.3B	30.4	3.0
4	multi30k_test_2018_flickr	compare	eng-ces+slk/...2022-03-13	33.4	facebook/nll...illed-1.3B	30.8	2.6
5	newssyscomb2009	compare	eng-ces+slk/...2022-03-13	25.3	facebook/nll...illed-1.3B	23.6	1.7
6	newstest2008	compare	eng-ces+slk/...2022-03-13	22.8	facebook/nllb-200-1.3B	21	1.8
7	newstest2009	compare	eng-ces+slk/...2022-03-13	24.3	facebook/nllb-200-3.3B	21.9	2.4
8	newstest2010	compare	eng-ces+slk/...2022-03-13	24.4	facebook/nll...illed-1.3B	22.5	1.9
9	newstest2011	compare	eng-ces+slk/...2022-03-13	25.5	facebook/nll...illed-1.3B	23.3	2.2
10	newstest2012	compare	eng-ces+slk/...2022-03-13	22.6	facebook/nllb-200-3.3B	20.9	1.7
11	newstest2013	compare	eng-ces+slk/...2022-03-13	27.4	facebook/nll...illed-1.3B	24.7	2.7
12	newstest2014	compare	eng-ces+slk/...2022-03-13	31.4	facebook/m2m100_1.2B	27.7	3.7
13	newstest2015	compare	eng-ces+slk/...2022-03-13	27	facebook/nll...illed-1.3B	24.4	2.6
14	newstest2016	compare	eng-ces+slk/...2022-03-13	29.9	facebook/m2m100_1.2B	27.1	2.8
15	newstest2017	compare	eng-ces+slk/...2022-03-13	24.9	facebook/m2m100_1.2B	22.7	2.2
16	newstest2018	compare	eng-ces+slk/...2022-03-13	24.6	facebook/m2m100_1.2B	22.3	2.3
17	newstest2019	compare	eng-ces+slk/...2022-03-13	26.4	facebook/nll...illed-1.3B	24.6	1.8
18	newstest2021	compare	deu+eng+fra+...2024-05-30	27.5	facebook/nll...illed-1.3B	25.2	2.3
19	ntrex128	compare	eng-ces+slk/...2022-03-13	31.3	facebook/nll...illed-1.3B	28.4	2.9
20	tatoeba-test-v2020-07-28	compare	en-cs/opus-2019-12-04	48.2	facebook/nllb-200-3.3B	45	3.2
21	tatoeba-test-v2021-03-30	compare	en-cs/opus-2019-12-04	48	facebook/nllb-200-3.3B	44.7	3.3
22	tatoeba-test-v2021-08-07	compare	en-cs/opus-2019-12-04	47.8	facebook/nllb-200-3.3B	44.6	3.2
		average		31.3		28.5	2.7