Nemotron-SEA-LION-v4.8-120B-A12B: benchmark results
Provider: NVIDIA. Released 2026-09-18. Access: Open.
Aggregate Rating 1626 ± 26, rank #789 of 10916 rated models, from 136 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| SEA-HELM (Malay) - Safety | 46.14 | Normalized Score | 98.2 |
| SEA-HELM (Filipino) - Kalahi (LLM Judge) | 76.62 | Normalized Score | 96.5 |
| SEA-HELM (Thai) - SEA-Safeguard | 75.64 | Normalized Score | 96.5 |
| SEA-HELM (Thai) - Causal Reasoning | 96.88 | Normalized Score | 93 |
| SEA-HELM (Thai) - SEA-MT-Bench (LLM Judge) | 84.19 | Normalized Score | 93 |
| SEA-HELM (English) | 66.19 | Normalized Score | 91.2 |
| SEA-HELM (English) - MMLU Redux | 91.21 | Normalized Score | 91.2 |
| SEA-HELM (Filipino) - Kalahi | 90.09 | Normalized Score | 91.2 |
| SEA-HELM (Malay) - SEA-Safeguard | 76.19 | Normalized Score | 91.2 |
| SEA-HELM (Malay) - Toxicity Detection | 16.09 | Normalized Score | 91.2 |
| SEA-HELM (Thai) - NLU | 63.36 | Normalized Score | 89.5 |
| SEA-HELM (Thai) - Sentiment Analysis | 40.4 | Normalized Score | 89.5 |
Interactive version: theaggregate.ai/model?slug=nemotron-sea-lion-v4-8-120b-a12b · How It Works · Data refreshed daily, snapshot 2026-11-08.