openthaigpt1.5-7B-instruct: benchmark results
Provider: Other. Access: Open.
Unified ELO 1564 ± 27, rank #778 of 2656 rated models, from 35 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Thai LLM NLU - wisesight_thai_sentiment_seacrowd_text | 50.24 | Accuracy (%) | 81.2 |
| ThaiSafetyBench - General Prompt Attacks | 5.59 | Attack Success Rate (%) | 78.3 |
| Thai LLM NLU - xnli.tha_seacrowd_pairs | 39.7 | Accuracy (%) | 76.8 |
| ThaiSafetyBench - Human-Chatbot Interaction Harms | 14.53 | Attack Success Rate (%) | 71.7 |
| Thai LLM - Extraction | 5.65 | Rating (0-10) | 69.7 |
| ThaiSafetyBench - Information Hazards | 2.2 | Attack Success Rate (%) | 67.4 |
| Thai LLM - Reasoning | 5.55 | Rating (0-10) | 66.9 |
| ThaiSafetyBench - Discrimination, Exclusion, Toxicity, Hateful, Offensive | 8.47 | Attack Success Rate (%) | 65.2 |
| Thai LLM MC - thaiexam_qa | 52.04 | Accuracy (%) | 64.5 |
| Thai LLM - Coding | 6.73 | Rating (0-10) | 63.4 |
| Thai LLM MC - m3exam_tha_seacrowd_qa | 54.01 | Accuracy (%) | 62.3 |
| Thai LLM NLG - xl_sum_tha_seacrowd_t2t (ROUGE1) | 27 | ROUGE1 | 61.2 |
Interactive version: theaggregate.ai/model?slug=openthaigpt1-5-7b-instruct · How It Works · Data refreshed daily, snapshot 2026-09-19.