Tsunami-1.0-7B-Instruct: benchmark results

Provider: Other. Access: Open.

Unified ELO 1533 ± 1, rank #524 of 1392 rated models, from 33 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open LLM Leaderboard - MATH Level 543.35Score92.4
Open LLM Leaderboard - IFEval73.09Score87.3
Open LLM Leaderboard - MuSR15.76Score83.8
Thai LLM NLU - xnli.tha_seacrowd_pairs41.04Accuracy (%)82.6
Open LLM Leaderboard - MMLU-Pro38.04Score82
Open LLM Leaderboard - BBH35.86Score72.9
Thai LLM NLG - xl_sum_tha_seacrowd_t2t (ROUGE2)10.14ROUGE271.6
Thai LLM NLG - xl_sum_tha_seacrowd_t2t (ROUGE1)29.1ROUGE170.1
Open LLM Leaderboard - GPQA8.39Score70
Thai LLM NLU - wisesight_thai_sentiment_seacrowd_text47.47Accuracy (%)69.6
Thai LLM NLG - xl_sum_tha_seacrowd_t2t (ROUGEL)19.44ROUGEL68.7
Thai LLM NLU - xcopa_tha_seacrowd_qa88.4Accuracy (%)67.4

Interactive version: theaggregate.ai/model?slug=tsunami-1-0-7b-instruct · How It Works · Data refreshed daily, snapshot 2026-09-05.