Skip to content

Benchmark

PashtoBench translation English to Pashto chrF++

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

Models measured
4
Scores recorded
4
Sources
1
IQ weight
No weight published

Score distribution

48.6
Mean
68.8
Median
21.9
Std dev
24.4
10th percentile
72.1
90th percentile
24.4
Min
24.472.1
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#ModelScore
1GPT 4odirection: en-pbt· PashtoBench v1.0 data docs72.1
2Claude Sonnet 5direction: en-pbt· PashtoBench v1.0 data docs68.8
3Llama 3.3 70B Instructdirection: en-pbt· PashtoBench v1.0 data docs29.2
4Qwen2.5 72B Instructdirection: en-pbt· PashtoBench v1.0 data docs24.4

Data credits: PashtoBench v1.0, Jawad Noori. Public results/scores numbers; code MIT and data/ CC BY 4.0 grants do not name results/, so no results licence is asserted.