Llama 3 8B Cpt Sahabatai V1 Base: benchmark results
Provider: Meta. Access: Open.
Aggregate Rating 1505 ± 46, rank #2261 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Sahabat-AI - Sundanese (SU) | 47.51 | Score (%) | 87.5 |
| Sahabat-AI - Javanese (JV) | 49.15 | Score (%) | 81.2 |
| Sahabat-AI - Indonesian Average | 47.78 | Average Score (%) | 68.8 |
| GoTo-AI Leaderboard - Balinese | 46.39 | Score (0-100) | 55.6 |
| GoTo-AI Leaderboard - Batak Toba | 30.75 | Score (0-100) | 51.9 |
| Sahabat-AI - Indonesian (ID) | 46.67 | Score (%) | 37.5 |
Interactive version: theaggregate.ai/model?slug=llama-3-8b-cpt-sahabatai-v1-base · How It Works · Data refreshed daily, snapshot 2026-11-08.