OPUS-MT Dashboard

Language pair: [eng-fra] all languages
Models: [all models] [OPUS-MT] [external] [compare]
Selected: Tatoeba-MT-models/iir-eng/opus4m-2020-08-12
Benchmark: all benchmarks [download]
Evaluation metric: bleu [spbleu][chrf][chrf++][comet]
Chart Type: [barchart] [heatmap]

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

	eng
asm	12.7
awa	8.1
ben	28.4
bho	6.5
guj	9.4
hin	25.2
kur	2.8
kur_Latn	2.8
mai	10.7
mar	18.3
npi	3.9
pan	14.7
pes	5.3
pus	3.0
rom	1.9
san	3.2
sin	8.8
snd	7.4
tgk	8.1
urd	18.6
zza	0.8

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