OPUS-MT Dashboard

Language pair: [eng-fra] all languages
Models: [all models] [OPUS-MT] [external] [compare]
Selected: Tatoeba-MT-models/sem-sem/opus-2020-09-26
Benchmark: all benchmarks [download]
Evaluation metric: bleu [spbleu][chrf][chrf++][comet]
Chart Type: [barchart] [heatmap]

Selected Benchmark: [avg] [flores200-devtest] [flores101-devtest] [tatoeba-test-v2021-03-30] [tico19-test] [tatoeba-test-v2020-07-28] [tatoeba-test-v2021-08-07]
Scores shown in percentage points

	acm	amh	apc	ara	arq	ary	arz	eng	heb	mlt	tir
acm	acm	1.9	0.4	44.3		0.4	4.5	19.4	7.6	8.7	1.9
amh	0.4	amh	0.4	4.6		0.3	1.1	23.0	4.7	5.3	2.6
apc	0.4	1.7	apc	14.9		0.4	2.9	19.3	6.9	7.6	1.9
ara	0.5	2.2	0.4	ara		0.4	6.5	32.4	25.4	11.4	2.3
arq					arq			7.5
ary	0.3	1.2	0.4	17.8		ary	2.4	11.8	4.5	5.3	1.2
arz	0.4	1.5	0.3	25.3		0.3	arz	15.2	5.9	6.5	1.6
eng	6.4	6.7	6.3	14.3	0.2	4.2	4.4	eng	26.4	22.1	3.9
heb	2.2	1.9	1.9	14.9		1.6	1.6	36.7	heb	8.9	1.9
mlt	1.2	3.2	1.3	11.8		0.9	4.7	46.5	12.4	mlt	2.7
tir	0.2	2.3	0.3	2.8		0.2	1.0	12.0	3.1	2.7	tir

color:
score:	0	10	20	30	40	50	60	70	80	90	100