SEA-HELM (Indonesian) - Natural Language Inference: leaderboard
Natural language inference in Indonesian: deciding whether a premise entails, contradicts or is neutral to a hypothesis. Normalized 0-100 score from AI Singapore's SEA-HELM leaderboard: (score - random baseline) / (maximum - baseline), floored at 0 and averaged over eight runs; higher is better.
Metric: Normalized Score. Source: leaderboard.sea-lion.ai. Saturation forecast: Estimated already saturated. 58 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3.6 27B (Reasoning) | 83.57 | #182 (Qwen 3.6 27B) |
| 2 | Qwen 3.5 27B (Reasoning) | 83.23 | #192 (Qwen 3.5 27B) |
| 3 | Qwen 3.5 122B A10B (Reasoning) | 83.11 | #184 (Qwen 3.5 122B A10B) |
| 4 | Qwen 3.6 35B A3B (Reasoning) | 82.7 | #238 (Qwen 3.6 35B A3B) |
| 5 | Gemma 4 26B A4B (IT) (Reasoning) | 82.56 | #293 (Gemma 4 26B A4B (IT)) |
| 6 | Qwen 3.5 9B (Reasoning) | 81.45 | #391 (Qwen 3.5 9B) |
| 7 | Qwen 3.5 35B A3B (Reasoning) | 81.38 | #268 (Qwen 3.5 35B A3B) |
| 8 | Qwen 3 VL 32B Instruct | 81.25 | #290 |
| 9 | Qwen 3.5 4B (Reasoning) | 80.52 | #514 (Qwen 3.5 4B) |
| 10 | Qwen 3 14B (Reasoning) | 79.82 | #542 (Qwen 3 14B) |
| 11 | GPT-OSS-120B (Reasoning) | 79.48 | #346 (GPT-OSS-120B) |
| 12 | Qwen 3 32B (Thinking) | 79.24 | #449 (Qwen 3 32B) |
| 13 | Llama-SEA-LION-v3-70B-IT | 78.43 | #407 |
| 14 | Llama 3.3 70B Instruct | 78.08 | #541 |
| 15 | Qwen 3 8B (Thinking) | 77.32 | #678 (Qwen 3 8B) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=sea-helm-indonesian-natural-language-inference · How It Works · Data refreshed daily, snapshot 2026-11-05.