SEA-HELM (Tamil) - SEA-MT-Bench (LLM Judge): leaderboard
Metric: Normalized Score. Source: leaderboard.sea-lion.ai. 44 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3.5 27B | 80.08 |
| 2 | Gemma 4 31B (IT) | 78.3 |
| 3 | Qwen 3.5 122B A10B | 77.93 |
| 4 | Gemma 4 26B A4B (IT) | 77.56 |
| 5 | Qwen 3.6 27B | 76.97 |
| 6 | Qwen 3.6 35B A3B | 75.74 |
| 7 | Gemma 4 12B (IT) | 75.72 |
| 8 | Qwen 3.5 35B A3B | 75.22 |
| 9 | Mistral Medium 3.5 | 73.32 |
| 10 | gemma-4-E4B-it | 71.81 |
| 11 | Qwen 3 VL 32B Instruct | 71.24 |
| 12 | Gemma 3 27B (IT) | 67.74 |
| 13 | Mistral Small 4 | 65.63 |
| 14 | Gemma 3 12B (IT) | 63.81 |
| 15 | Qwen 3.5 9B | 63.54 |
Interactive version: theaggregate.ai/benchmark?slug=sea-helm-tamil-sea-mt-bench-llm-judge · How It Works · Data refreshed daily, snapshot 2026-09-19.