Llama-SEA-LION-v3-8B-IT: benchmark results
Provider: Meta. Access: Open.
Unified ELO 1495 ± 19, rank #793 of 1632 rated models, from 143 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| SEA-HELM (Vietnamese) - Sentiment Analysis | 63.88 | Normalized Score | 93 |
| SEA-HELM (Indonesian) - SEA-Safeguard | 73.74 | Normalized Score | 91.2 |
| SEA-HELM (Thai) - Summarization | 23.93 | Normalized Score | 87.7 |
| SEA-HELM (Indonesian) - Summarization | 17.15 | Normalized Score | 80.7 |
| SEA-HELM (Malay) - Toxicity Detection | 14.39 | Normalized Score | 80.7 |
| SEA-HELM (Tamil) - Summarization | 11.46 | Normalized Score | 71.9 |
| SEA-HELM (Malay) - Safety | 41.15 | Normalized Score | 70.2 |
| SEA-HELM (Vietnamese) - Summarization | 16.3 | Normalized Score | 66.7 |
| ThaiSafetyBench - Malicious Uses | 10.29 | Attack Success Rate (%) | 65.2 |
| SEA-HELM (Vietnamese) - NLG | 53.93 | Normalized Score | 64.9 |
| SEA-HELM (Indonesian) - NLG | 54.5 | Normalized Score | 63.2 |
| SEA-HELM (Malay) - SEA-NLI | 57.17 | Normalized Score | 63.2 |
Interactive version: theaggregate.ai/model?slug=llama-sea-lion-v3-8b-it · How It Works · Data refreshed daily, snapshot 2026-10-08.