Benchmark
PashtoBench factual QA exact match (Pashto)
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
- 0.006
- Mean
- 0
- Median
- 0.011
- Std dev
- 0
- 10th percentile
- 0.025
- 90th percentile
- 0
- Min
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | Llama 3.3 70B Instructcolumn: EM pbt· PashtoBench v1.0 data docs | 0.025 |
| 2 | Claude Sonnet 5column: EM pbt· PashtoBench v1.0 data docs | 0 |
| 3 | GPT 4ocolumn: EM pbt· PashtoBench v1.0 data docs | 0 |
| 4 | Qwen2.5 72B Instructcolumn: EM pbt· PashtoBench v1.0 data docs | 0 |