SEA-HELM (Filipino) - Kalahi: leaderboard
Kalahi: multiple-choice questions on culturally appropriate responses in Filipino situations. 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 | Gemma 4 26B A4B (IT) (Reasoning) | 91.35 | #293 (Gemma 4 26B A4B (IT)) |
| 2 | Qwen 3.5 122B A10B (Reasoning) | 91.28 | #184 (Qwen 3.5 122B A10B) |
| 3 | Qwen 3.6 35B A3B (Reasoning) | 90.16 | #238 (Qwen 3.6 35B A3B) |
| 4 | Qwen 3.5 35B A3B (Reasoning) | 88.54 | #268 (Qwen 3.5 35B A3B) |
| 5 | Llama-SEA-LION-v3-70B-IT | 87.88 | #407 |
| 6 | Llama 4 Scout Instruct | 85.98 | #547 |
| 7 | Qwen 3.5 27B (Reasoning) | 85.69 | #192 (Qwen 3.5 27B) |
| 8 | Qwen 3.6 27B (Reasoning) | 85.14 | #182 (Qwen 3.6 27B) |
| 9 | GPT-OSS-120B (Reasoning) | 83.16 | #346 (GPT-OSS-120B) |
| 10 | Llama 3.3 70B Instruct | 82.97 | #541 |
| 11 | Gemma 3 27B (IT) | 82.94 | #528 |
| 12 | Gemma 3 12B (IT) | 81.24 | #672 |
| 13 | Qwen 3 VL 32B Instruct | 80.41 | #290 |
| 14 | Qwen 3 32B (Thinking) | 78.63 | #449 (Qwen 3 32B) |
| 15 | Mistral Small 4 | 77.3 | #522 |
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-filipino-kalahi · How It Works · Data refreshed daily, snapshot 2026-11-05.