SEA-HELM (Indonesian) - Question Answering: leaderboard
Metric: Normalized Score. Source: leaderboard.sea-lion.ai. 44 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 4 Scout Instruct | 83.9 |
| 2 | Gemma 4 31B (IT) | 83.79 |
| 3 | Qwen 3.5 27B | 83.22 |
| 4 | Qwen 3.6 27B | 82.74 |
| 5 | Qwen 3.6 35B A3B | 82.54 |
| 6 | Qwen 3.5 122B A10B | 82.52 |
| 7 | Gemma 3 27B (IT) | 81.92 |
| 8 | Gemma 3 12B (IT) | 81.26 |
| 9 | Qwen 3 VL 8B Instruct | 80.8 |
| 10 | Gemma 4 26B A4B (IT) | 80.69 |
| 11 | Qwen 3.5 35B A3B | 80.27 |
| 12 | Mistral Medium 3.5 | 80.04 |
| 13 | Llama 3.3 70B Instruct | 79.69 |
| 14 | Gemma 4 12B (IT) | 78.7 |
| 15 | Mistral Small 4 | 77.94 |
Interactive version: theaggregate.ai/benchmark?slug=sea-helm-indonesian-question-answering · How It Works · Data refreshed daily, snapshot 2026-09-19.