SEA-HELM (Vietnamese) - Natural Language Inference: leaderboard
Natural language inference in Vietnamese: 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: Around February 2027. 58 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3.6 27B (Reasoning) | 63.2 | #182 (Qwen 3.6 27B) |
| 2 | Qwen 3.6 35B A3B (Reasoning) | 62.92 | #238 (Qwen 3.6 35B A3B) |
| 3 | Llama 3.3 70B Instruct | 60.23 | #541 |
| 4 | Qwen 3.5 35B A3B (Reasoning) | 59.13 | #268 (Qwen 3.5 35B A3B) |
| 5 | Qwen 3.5 27B (Reasoning) | 59.11 | #192 (Qwen 3.5 27B) |
| 6 | Gemma 4 26B A4B (IT) (Reasoning) | 57.96 | #293 (Gemma 4 26B A4B (IT)) |
| 7 | Qwen 3.5 122B A10B (Reasoning) | 57.91 | #184 (Qwen 3.5 122B A10B) |
| 8 | Gemma 3 12B (IT) | 57.67 | #672 |
| 9 | Gemma 3 27B (IT) | 55.94 | #528 |
| 10 | Llama-SEA-LION-v3-70B-IT | 55.02 | #407 |
| 11 | GPT-OSS-120B (Reasoning) | 54.58 | #346 (GPT-OSS-120B) |
| 12 | Qwen 3.5 9B (Reasoning) | 54.38 | #391 (Qwen 3.5 9B) |
| 13 | Qwen 3.5 4B (Reasoning) | 52.58 | #514 (Qwen 3.5 4B) |
| 14 | Qwen 3 32B (Thinking) | 51.51 | #449 (Qwen 3 32B) |
| 15 | GLM-4.7 Flash (Reasoning) | 51.33 | #543 (GLM-4.7 Flash) |
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-vietnamese-natural-language-inference · How It Works · Data refreshed daily, snapshot 2026-11-05.