Qwen-SEA-LION-v5-397B-A17B-Plus: benchmark results
AI Singapore's SEA-LION v5 flagship (Nov 18, 2026, MIT), converting Qwen3.5-397B-A17B to Qwen4Exp with ~164B host-offloaded n-gram embedding tables (569.45B safetensors total, 17B active). Provider: AI Singapore. Released 2026-11-18. Access: Open.
Aggregate Rating 1751 ± 36, rank #178 of 10524 rated models, from 132 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| SEA-HELM | 77.1 | Mean Score (%) | 100 |
| SEA-HELM (Burmese) | 70.58 | Normalized Score | 100 |
| SEA-HELM (Burmese) - Belebele QA | 88.01 | Normalized Score | 100 |
| SEA-HELM (Burmese) - Causal Reasoning | 84.77 | Normalized Score | 100 |
| SEA-HELM (Burmese) - Global MMLU Lite | 81.72 | Normalized Score | 100 |
| SEA-HELM (Burmese) - NLR | 70.54 | Normalized Score | 100 |
| SEA-HELM (Burmese) - NLU | 68.09 | Normalized Score | 100 |
| SEA-HELM (Filipino) - Global MMLU Lite | 87.48 | Normalized Score | 100 |
| SEA-HELM (Filipino) - Sentiment Analysis | 72.17 | Normalized Score | 100 |
| SEA-HELM (Indonesian) | 80.29 | Normalized Score | 100 |
| SEA-HELM (Malay) - Safety | 47.68 | Normalized Score | 100 |
| SEA-HELM (Malay) - Sentiment Analysis | 53.52 | Normalized Score | 100 |
Interactive version: theaggregate.ai/model?slug=qwen-sea-lion-v5-397b-a17b-plus · How It Works · Data refreshed daily, snapshot 2026-11-22.