openthaigpt-1.0.0-70B-chat: benchmark results
Provider: Other. Access: Open.
Aggregate Rating 1423 ± 25, rank #5031 of 10916 rated models, from 33 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Thai LLM NLG - flores200_eng_Latn_tha_Thai_seacrowd_t2t (SacreBLEU) | 38.9 | SacreBLEU | 89.6 |
| Thai LLM NLG - flores200_eng_Latn_tha_Thai_seacrowd_t2t (BLEU) | 27.72 | BLEU | 88.1 |
| Open LLM Leaderboard v1 - MMLU | 65.53 | Accuracy (%) (5-shot) | 84.8 |
| Thai LLM NLG - flores200_eng_Latn_tha_Thai_seacrowd_t2t (chrF++) | 53.54 | chrF++ | 82.1 |
| Thai LLM NLG - iapp_squad_seacrowd_qa (ROUGE1) | 62.17 | ROUGE1 | 82.1 |
| Thai LLM NLG - iapp_squad_seacrowd_qa (ROUGEL) | 61.68 | ROUGEL | 82.1 |
| Thai LLM NLG - xl_sum_tha_seacrowd_t2t (ROUGEL) | 21.18 | ROUGEL | 82.1 |
| Thai LLM NLG - iapp_squad_seacrowd_qa (ROUGE2) | 46.68 | ROUGE2 | 79.1 |
| Thai LLM NLG - xl_sum_tha_seacrowd_t2t (ROUGE1) | 29.17 | ROUGE1 | 73.1 |
| Open LLM Leaderboard v1 - HellaSwag | 85.08 | Normalized accuracy (%) (10-shot) | 71.6 |
| Open LLM Leaderboard v1 - ARC Challenge | 64.42 | Normalized accuracy (%) (25-shot) | 66.4 |
| Thai LLM NLG - xl_sum_tha_seacrowd_t2t (ROUGE2) | 9.69 | ROUGE2 | 65.7 |
Interactive version: theaggregate.ai/model?slug=openthaigpt-1-0-0-70b-chat · How It Works · Data refreshed daily, snapshot 2026-11-08.