Thai LLM NLU - wisesight_thai_sentiment_seacrowd_text: leaderboard
The Thai LLM Leaderboard, run as an open results dataset by the Thai LLM community. Natural-language understanding accuracy on Wisesight Thai sentiment classification (SEACrowd harness).
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 70 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 2 72B Instruct | 60.09 | #576 |
| 2 | GPT-4o (2024-05-13) | 56.76 | #325 |
| 3 | DeepSeek R1 Distill Llama 70B | 56.76 | #594 |
| 4 | Qwen 2.5 7B Instruct | 52.86 | #862 |
| 5 | Qwen 2.5 14B Instruct | 52.68 | #647 |
| 6 | GPT-4o Mini (2024-07-18) | 52.3 | #558 |
| 7 | Llama 3.3 70B Instruct | 51.82 | #534 |
| 8 | Llama 3.1 Nemotron 70B Instruct | 51.03 | #554 |
| 9 | Llama 3.1 8B CPT SEA LIONv3 Instruct | 49.79 | #810 |
| 10 | Llama 3.1 70B Cpt Sea Lionv3 Instruct | 49.72 | #548 |
| 11 | Claude 3.5 Sonnet (20240620) | 49.61 | #299 |
| 12 | Llama 3.1 70B Instruct | 49.53 | #561 |
| 13 | Gemma 2 2B (IT) | 47.66 | #1341 |
| 14 | Qwen 2 7B Instruct | 47.06 | #991 |
| 15 | Qwen 2.5 3B Instruct | 46.76 | #1162 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=thai-llm-nlu-wisesight-thai-sentiment-seacrowd-text · How It Works · Data refreshed daily, snapshot 2026-10-20.