SEA-HELM (Indonesian) - Causal Reasoning: leaderboard
Causal commonsense reasoning in Indonesian: choosing the more plausible cause or effect of a described situation. 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 3.5 122B A10B (Reasoning) | 97.04 | #184 (Qwen 3.5 122B A10B) |
| 2 | Qwen 3.6 35B A3B (Reasoning) | 96.47 | #238 (Qwen 3.6 35B A3B) |
| 3 | Qwen 3.5 27B (Reasoning) | 96.41 | #192 (Qwen 3.5 27B) |
| 4 | Gemma 4 26B A4B (IT) (Reasoning) | 96.4 | #293 (Gemma 4 26B A4B (IT)) |
| 5 | Qwen 3.6 27B (Reasoning) | 96.35 | #182 (Qwen 3.6 27B) |
| 6 | Qwen 3.5 35B A3B (Reasoning) | 95.87 | #268 (Qwen 3.5 35B A3B) |
| 7 | Qwen 3 32B (Thinking) | 95.49 | #449 (Qwen 3 32B) |
| 8 | Qwen 3.5 9B (Reasoning) | 94.86 | #391 (Qwen 3.5 9B) |
| 9 | Qwen 3 14B (Reasoning) | 93.84 | #542 (Qwen 3 14B) |
| 10 | Qwen 3 VL 32B Instruct | 93.6 | #290 |
| 11 | Llama 3.3 70B Instruct | 93.07 | #541 |
| 12 | Llama-SEA-LION-v3-70B-IT | 92.1 | #407 |
| 13 | GPT-OSS-120B (Reasoning) | 91.91 | #346 (GPT-OSS-120B) |
| 14 | Qwen 3.5 4B (Reasoning) | 90.75 | #514 (Qwen 3.5 4B) |
| 15 | Llama 4 Scout Instruct | 90.63 | #547 |
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-causal-reasoning · How It Works · Data refreshed daily, snapshot 2026-11-05.