Thai LLM MC - m3exam_tha_seacrowd_qa: leaderboard
Metric: Accuracy (%). Source: huggingface.co. 70 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4o (2024-05-13) | 68.63 |
| 2 | Claude 3.5 Sonnet (20240620) | 67.62 |
| 3 | Qwen 2.5 72B Instruct | 66.33 |
| 4 | Qwen 2.5 32B Instruct | 63.7 |
| 5 | Llama 3.1 70B Cpt Sea Lionv3 Instruct | 63.1 |
| 6 | Qwen 2 72B Instruct | 63.05 |
| 7 | Tsunami-1.0-14B-Instruct | 63.05 |
| 8 | Llama 3.3 70B Instruct | 62.22 |
| 9 | QwQ 32B-Preview | 61.3 |
| 10 | Llama 3.1 Nemotron 70B Instruct | 60.79 |
| 11 | Llama 3.1 70B Instruct | 60.52 |
| 12 | Qwen 2.5 14B Instruct | 59.55 |
| 13 | DeepSeek R1 Distill Qwen 32B | 58.49 |
| 14 | GPT-4o Mini (2024-07-18) | 57.84 |
| 15 | gemma2-9B-cpt-sea-lionv3-instruct | 56 |
Interactive version: theaggregate.ai/benchmark?slug=thai-llm-mc-m3exam-tha-seacrowd-qa · How It Works · Data refreshed daily, snapshot 2026-09-05.