Benchmark
LaoBench Lao-7k translation Society & Law chrF++ (zero-shot)
Multilingual benchmarks compare model performance across languages. Higher scores are better.
- Models measured
- 3
- Scores recorded
- 3
- Sources
- 1
- IQ weight
- No weight published
Score distribution
- 51.8
- Mean
- 52.4
- Median
- 1.1
- Std dev
- 50.3
- 10th percentile
- 52.8
- 90th percentile
- 50.3
- Min
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Protable: 7, mode: single· LaoBench (arXiv:2511.11334v3) data | 52.8 |
| 2 | Qwen3 Maxtable: 7, mode: single· LaoBench (arXiv:2511.11334v3) data | 52.4 |
| 3 | GPT 5table: 7, mode: single· LaoBench (arXiv:2511.11334v3) data | 50.3 |