SEA-HELM (Indonesian) - Syntax (LLM Judge): leaderboard
LINDSEA syntax items in Indonesian whose responses are graded against criteria by an LLM judge. 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: Around February 2027. 58 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemma 4 26B A4B (IT) (Reasoning) | 85.65 | #293 (Gemma 4 26B A4B (IT)) |
| 2 | Qwen 3.5 35B A3B (Reasoning) | 79.44 | #268 (Qwen 3.5 35B A3B) |
| 3 | Qwen 3.5 122B A10B (Reasoning) | 78.8 | #184 (Qwen 3.5 122B A10B) |
| 4 | Qwen 3.6 27B (Reasoning) | 77.6 | #182 (Qwen 3.6 27B) |
| 5 | Qwen 3.6 35B A3B (Reasoning) | 76.44 | #238 (Qwen 3.6 35B A3B) |
| 6 | GPT-OSS-120B (Reasoning) | 73.99 | #346 (GPT-OSS-120B) |
| 7 | Qwen 3.5 27B (Reasoning) | 73.04 | #192 (Qwen 3.5 27B) |
| 8 | Qwen 3.5 9B (Reasoning) | 70.37 | #391 (Qwen 3.5 9B) |
| 9 | Qwen 3 32B (Thinking) | 67.02 | #449 (Qwen 3 32B) |
| 10 | Qwen 3.5 4B (Reasoning) | 65.14 | #514 (Qwen 3.5 4B) |
| 11 | Qwen 3 14B (Reasoning) | 64.61 | #542 (Qwen 3 14B) |
| 12 | Gemma 3 27B (IT) | 63.02 | #528 |
| 13 | Qwen 3 8B (Thinking) | 60.25 | #678 (Qwen 3 8B) |
| 14 | Llama 3.3 70B Instruct | 59.2 | #541 |
| 15 | Llama-SEA-LION-v3-70B-IT | 59.1 | #407 |
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-syntax-llm-judge · How It Works · Data refreshed daily, snapshot 2026-11-05.