SEA-HELM (Filipino) - Summarization: leaderboard
Abstractive summarization of Filipino news articles. 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 2037. 58 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemma 4 26B A4B (IT) (Reasoning) | 26.03 | #293 (Gemma 4 26B A4B (IT)) |
| 2 | Llama 3.1 8B Instruct | 25.52 | #1017 |
| 3 | Llama-SEA-LION-v3-70B-IT | 25.04 | #407 |
| 4 | Llama 3.3 70B Instruct | 24.33 | #541 |
| 5 | Qwen 3.5 122B A10B (Reasoning) | 22.84 | #184 (Qwen 3.5 122B A10B) |
| 6 | GLM-4.7 Flash (Reasoning) | 22.53 | #543 (GLM-4.7 Flash) |
| 7 | Qwen 3.5 27B (Reasoning) | 22.44 | #192 (Qwen 3.5 27B) |
| 8 | MERaLiON-2-10B | 21.67 | #600 |
| 9 | Qwen 3.5 9B (Reasoning) | 21.46 | #391 (Qwen 3.5 9B) |
| 10 | Llama 4 Scout Instruct | 21.36 | #547 |
| 11 | Qwen 3.5 35B A3B (Reasoning) | 21.21 | #268 (Qwen 3.5 35B A3B) |
| 12 | OLMo 3.1 32B Instruct | 21.15 | #755 |
| 13 | Qwen 3.6 27B (Reasoning) | 21.03 | #182 (Qwen 3.6 27B) |
| 14 | Qwen 3.6 35B A3B (Reasoning) | 21 | #238 (Qwen 3.6 35B A3B) |
| 15 | Qwen 3 VL 8B Instruct | 20.8 | #419 |
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-summarization · How It Works · Data refreshed daily, snapshot 2026-11-05.