Thai LLM - Reasoning: leaderboard
Metric: Rating (0-10). Source: huggingface.co. 72 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4o (2024-05-13) | 9 |
| 2 | Gemini 1.5 Pro (001) | 7.95 |
| 3 | Claude 3.5 Sonnet (20240620) | 7.55 |
| 4 | Qwen 2.5 72B Instruct | 7.15 |
| 5 | DeepSeek R1 Distill Llama 70B | 7.15 |
| 6 | DeepSeek R1 Distill Qwen 32B | 6.8 |
| 7 | GPT-4o Mini (2024-07-18) | 6.7 |
| 8 | Qwen 2.5 32B Instruct | 6.65 |
| 9 | Qwen 2.5 14B Instruct | 6.6 |
| 10 | Llama 3.3 70B Instruct | 6.5 |
| 11 | Qwen 2 72B Instruct | 6.5 |
| 12 | Llama 3.1 Nemotron 70B Instruct | 6.45 |
| 13 | Llama 3.1 70B Cpt Sea Lionv3 Instruct | 6.35 |
| 14 | Gemma 2 9B (IT) | 5.9 |
| 15 | Llama 3.1 70B Instruct | 5.85 |
Interactive version: theaggregate.ai/benchmark?slug=thai-llm-reasoning · How It Works · Data refreshed daily, snapshot 2026-09-05.