Skip to content

Benchmark

Persian-family QA Exact Match (Iranian Persian aggregate, zero-shot)

Multilingual benchmarks compare model performance across languages. Scores are percentages. Higher is better.

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

Score distribution

8.4%
Mean
10.8%
Median
2.4%
Std dev
6.1%
10th percentile
10.8%
90th percentile
6.1%
Min
6.1%10.8%
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#ModelScore
1Qwen2.5 1.5B Instructtable: 2, task: QA, metric: Exact Match, subset: Iranian Persian aggregate· Persian, Dari and Tajiki paper data docs10.8%
2Qwen2.5 3B Instructtable: 2, task: QA, metric: Exact Match, subset: Iranian Persian aggregate· Persian, Dari and Tajiki paper data docs6.1%

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)