Gemma2-9B-WangchanLIONv2-instruct: benchmark results

Provider: Other.

Aggregate Rating 1431 ± 43, rank #4797 of 10516 rated models, from 27 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Thai LLM NLG - iapp_squad_seacrowd_qa (ROUGE1)82.97ROUGE1100
Thai LLM NLG - iapp_squad_seacrowd_qa (ROUGE2)71.21ROUGE2100
Thai LLM NLG - iapp_squad_seacrowd_qa (ROUGEL)82.87ROUGEL100
Thai LLM NLG - xl_sum_tha_seacrowd_t2t (ROUGE2)16.54ROUGE2100
Thai LLM NLG - xl_sum_tha_seacrowd_t2t (ROUGE1)37.43ROUGE198.5
Thai LLM NLG - xl_sum_tha_seacrowd_t2t (ROUGEL)27.99ROUGEL98.5
Thai LLM NLG - flores200_tha_Thai_eng_Latn_seacrowd_t2t (BLEU)52.6BLEU91
Thai LLM NLG - flores200_tha_Thai_eng_Latn_seacrowd_t2t (SacreBLEU)36.59SacreBLEU91
Thai LLM NLU - xnli.tha_seacrowd_pairs41.48Accuracy (%)85.5
Thai LLM NLG - flores200_tha_Thai_eng_Latn_seacrowd_t2t (chrF++)59.44chrF++79.1
Thai LLM NLG - flores200_eng_Latn_tha_Thai_seacrowd_t2t (BLEU)25.85BLEU74.6
Thai LLM NLG - flores200_eng_Latn_tha_Thai_seacrowd_t2t (SacreBLEU)36.83SacreBLEU74.6

Interactive version: theaggregate.ai/model?slug=gemma2-9b-wangchanlionv2-instruct · How It Works · Data refreshed daily, snapshot 2026-11-21.