Benchmark
PashtoBench translation Pashto to English BLEU
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
- 38.1
- Mean
- 70
- Median
- 34.3
- Std dev
- 2.3
- 10th percentile
- 74.6
- 90th percentile
- 2.3
- Min
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 5direction: pbt-en· PashtoBench v1.0 data docs | 74.6 |
| 2 | GPT 4odirection: pbt-en· PashtoBench v1.0 data docs | 70 |
| 3 | Qwen2.5 72B Instructdirection: pbt-en· PashtoBench v1.0 data docs | 5.3 |
| 4 | Llama 3.3 70B Instructdirection: pbt-en· PashtoBench v1.0 data docs | 2.3 |