SEA-HELM (Tamil) - Sentiment Analysis: leaderboard
Metric: Normalized Score. Source: leaderboard.sea-lion.ai. 44 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemma 3 27B (IT) | 97.31 |
| 2 | Gemma 4 31B (IT) | 96.77 |
| 3 | Llama 4 Scout Instruct | 96.69 |
| 4 | Mistral Medium 3.5 | 96.29 |
| 5 | Qwen 3.5 122B A10B | 96.12 |
| 6 | Gemma 4 26B A4B (IT) | 95.82 |
| 7 | Qwen 3.5 27B | 95.75 |
| 8 | Llama 3.3 70B Instruct | 95.67 |
| 9 | gemma-4-E4B-it | 95.57 |
| 10 | Gemma 4 E2B (IT) | 95.32 |
| 11 | Gemma 4 12B (IT) | 95.11 |
| 12 | Gemma 3 12B (IT) | 94.83 |
| 13 | Qwen 3.6 35B A3B | 94.25 |
| 14 | Qwen 3 VL 32B Instruct | 94.06 |
| 15 | Qwen 3.5 35B A3B | 93.93 |
Interactive version: theaggregate.ai/benchmark?slug=sea-helm-tamil-sentiment-analysis · How It Works · Data refreshed daily, snapshot 2026-09-19.