Thai LLM NLU - belebele_tha_thai_seacrowd_qa: leaderboard

Metric: Accuracy (%). Source: huggingface.co. 70 models tracked.

Top models

#ModelScore
1Claude 3.5 Sonnet (20240620)90.22
2GPT-4o (2024-05-13)89.78
3Llama 3.1 Nemotron 70B Instruct88.33
4Qwen 2.5 32B Instruct88.22
5Qwen 2.5 72B Instruct87.67
6Llama 3.1 70B Instruct87.56
7Llama 3.1 70B Cpt Sea Lionv3 Instruct87.22
8Llama 3.3 70B Instruct86.67
9Qwen 2 72B Instruct86.11
10Qwen 2.5 14B Instruct85.44
11Tsunami-1.0-14B-Instruct85.33
12gemma2-9B-cpt-sea-lionv3-instruct85
13GPT-4o Mini (2024-07-18)84.78
14QwQ 32B-Preview84
15Gemma 2 9B (IT)83.78

Interactive version: theaggregate.ai/benchmark?slug=thai-llm-nlu-belebele-tha-thai-seacrowd-qa · How It Works · Data refreshed daily, snapshot 2026-09-05.