HELM ThaiExam - TPAT-1: leaderboard
HELM ThaiExam TPAT-1 professional-aptitude test in Thai.
Metric: EM. Source: crfm.stanford.edu. Saturation forecast: Estimated already saturated. 42 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Llama 3.1 405B Instruct | 68.1 | #457 |
| 2 | Claude 3.5 Sonnet (20240620) | 64.66 | #299 |
| 3 | Llama 3.1 70B Instruct | 59.48 | #561 |
| 4 | Claude 3 Opus (20240229) | 59.48 | #412 |
| 5 | Gemini 1.5 Pro (001) | 59.48 | #444 |
| 6 | Qwen 2 72B Instruct | 55.17 | #576 |
| 7 | GPT-4o (2024-05-13) | 54.31 | #325 |
| 8 | Qwen 1.5 72B Chat | 54.31 | #776 |
| 9 | Gemini 1.5 Flash (001) | 53.45 | #577 |
| 10 | Llama 3 70B Instruct | 52.59 | #634 |
| 11 | Claude 3 Sonnet (20240229) | 50.86 | #677 |
| 12 | GPT-4 Turbo | 49.14 | #404 |
| 13 | GPT-4o Mini (2024-07-18) | 47.41 | #558 |
| 14 | Claude 3 Haiku (20240307) | 47.41 | #771 |
| 15 | Command-R+ | 47.41 | #891 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=helm-thaiexam-tpat-1 · How It Works · Data refreshed daily, snapshot 2026-10-20.