mistral-7B-test-v0.3: benchmark results
Provider: Mistral. Access: Open.
Unified ELO 1375 ± 17, rank #2088 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - KorNAT-Harmless | 63.66 | Normalized accuracy (%) | 65.3 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 19.64 | Exact match, flexible extract (%) | 51 |
| Open Korean LLM - Ko-GPQA-Diamond | 23.23 | Normalized accuracy (%) | 48.1 |
| Open Korean LLM - Ko-Winogrande | 58.96 | Accuracy (%) | 43 |
| Open Korean LLM - KorNAT-Helpful | 46.08 | Normalized accuracy (%) | 42.1 |
| Open Korean LLM Leaderboard | 31.66 | Average Score (%) | 28.8 |
| Open Ko-LLM Leaderboard | 31.66 | Average (%) | 28.6 |
| Open Korean LLM - Ko-EQ-Bench | 0 | EQ-Bench score | 24.5 |
| Open Korean LLM - Ko-IFEval | 18.67 | Strict accuracy, prompt/instruction mean (%) | 14.3 |
| Open Korean LLM - KorNAT-CKA | 16.36 | Normalized accuracy (%) | 12 |
Interactive version: theaggregate.ai/model?slug=mistral-7b-test-v0-3 · How It Works · Data refreshed daily, snapshot 2026-09-19.