komt-mistral-7B-v1: benchmark results

Provider: Other. Access: Open.

Unified ELO 1394 ± 17, rank #1955 of 2656 rated models, from 10 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open Korean LLM - Ko-IFEval36.57Strict accuracy, prompt/instruction mean (%)75.8
Open Korean LLM - Ko-EQ-Bench7.86EQ-Bench score54.4
Open Korean LLM - KorNAT-CKA23.35Normalized accuracy (%)44.6
Open Korean LLM - Ko-GSM8K (flexible extract)14.63Exact match, flexible extract (%)43.3
Open Korean LLM Leaderboard33.14Average Score (%)38.5
Open Ko-LLM Leaderboard33.14Average (%)38.3
Open Korean LLM - KorNAT-Harmless61.8Normalized accuracy (%)34.1
Open Korean LLM - KorNAT-Helpful43.97Normalized accuracy (%)26.7
Open Korean LLM - Ko-Winogrande54.62Accuracy (%)16.8
Open Korean LLM - Ko-GPQA-Diamond19.19Normalized accuracy (%)6.4

Interactive version: theaggregate.ai/model?slug=komt-mistral-7b-v1 · How It Works · Data refreshed daily, snapshot 2026-09-19.