Gemma-SEA-LION-v4-27B-IT: benchmark results
Provider: Google. Access: API.
Aggregate Rating 1599 ± 30, rank #1023 of 10916 rated models, from 130 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| SEA-HELM (Filipino) - Safety | 68.75 | Normalized Score | 98.2 |
| SEA-HELM (Filipino) - Toxicity Detection | 61.48 | Normalized Score | 98.2 |
| SEA-HELM (Malay) - SEA-Safeguard | 77.21 | Normalized Score | 96.5 |
| SEA-HELM (Malay) - Safety | 45.86 | Normalized Score | 96.5 |
| SEA-HELM (Indonesian) - SEA-Safeguard | 75.25 | Normalized Score | 94.7 |
| SEA-HELM (Tamil) - Sentiment Analysis | 97.21 | Normalized Score | 94.7 |
| SEA-HELM (Filipino) - Translations | 91.25 | Normalized Score | 93 |
| SEA-HELM (Tamil) - Natural Language Inference | 54.59 | Normalized Score | 93 |
| SEA-HELM (Burmese) - Sentiment Analysis | 46.56 | Normalized Score | 91.2 |
| SEA-HELM (Tamil) - NLU | 82.5 | Normalized Score | 91.2 |
| SEA-HELM (Filipino) - SEA-Safeguard | 76.03 | Normalized Score | 89.5 |
| SEA-HELM (Indonesian) - Safety | 61.78 | Normalized Score | 89.5 |
Interactive version: theaggregate.ai/model?slug=gemma-sea-lion-v4-27b-it · How It Works · Data refreshed daily, snapshot 2026-11-08.