Astral-7B-Instruct-v0.01: benchmark results
Provider: Other. Access: Open.
Unified ELO 1354 ± 17, rank #2209 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-GPQA-Diamond | 24.24 | Normalized accuracy (%) | 62.4 |
| Open Korean LLM - KorNAT-CKA | 25.5 | Normalized accuracy (%) | 58.6 |
| Open Korean LLM - Ko-IFEval | 27.33 | Strict accuracy, prompt/instruction mean (%) | 42.7 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 13.12 | Exact match, flexible extract (%) | 40.1 |
| Open Korean LLM - Ko-Winogrande | 57.22 | Accuracy (%) | 30.4 |
| Open Korean LLM Leaderboard | 30.37 | Average Score (%) | 18.6 |
| Open Ko-LLM Leaderboard | 30.37 | Average (%) | 18.5 |
| Open Korean LLM - KorNAT-Harmless | 60.41 | Normalized accuracy (%) | 14.4 |
| Open Korean LLM - KorNAT-Helpful | 40.12 | Normalized accuracy (%) | 6.8 |
| Open Korean LLM - Ko-EQ-Bench | -13.54 | EQ-Bench score | 1.7 |
Interactive version: theaggregate.ai/model?slug=astral-7b-instruct-v0-01 · How It Works · Data refreshed daily, snapshot 2026-09-19.