Skip to content

Benchmark

Persian-family Machine Translation BLEU (Persian/Dari/Tajiki pooled, all sources, zero-shot)

Multilingual benchmarks compare model performance across languages. Higher scores are better.

Models measured
2
Scores recorded
2
Sources
1
IQ weight
No weight published

Score distribution

6.1
Mean
7.6
Median
1.5
Std dev
4.6
10th percentile
7.6
90th percentile
4.6
Min
4.67.6
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#ModelScore
1Qwen2.5 3B Instructtable: 2, task: Machine Translation, metric: BLEU, subset: Persian/Dari/Tajiki pooled, all sources· Persian, Dari and Tajiki paper data docs7.6
2Qwen2.5 1.5B Instructtable: 2, task: Machine Translation, metric: BLEU, subset: Persian/Dari/Tajiki pooled, all sources· Persian, Dari and Tajiki paper data docs4.6

Data credits: Noor Mairukh Khan Arnob and Abu Bakar Siddique Mahi. One Language, Three of Its Voices, SilkRoadNLP 2026, ©2026 ACL, CC BY 4.0. Numeric/protocol extract from Tables 2, 5 and 6; language and pooled subsets retained. (licence)