Configurable-Llama-3.1-8B-Instruct: benchmark results
vicgalle's configurable-safety-tuning research fine-tune of Llama 3.1 8B Instruct, letting system prompts set safety behavior at inference time. Provider: Other. Released 2024-07-24. Access: Open.
Unified ELO 1521 ± 1, rank #594 of 1392 rated models, from 11 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard - IFEval | 83.12 | Score | 99.3 |
| Open Korean LLM Leaderboard | 45.5 | Average Score (%) | 91.7 |
| Open LLM Leaderboard - MATH Level 5 | 17.3 | Score | 66.1 |
| Open LLM Leaderboard - MMLU-Pro | 28.8 | Score | 54.5 |
| Open LLM Leaderboard - BBH | 29.66 | Score | 50.4 |
| UGI - Natural Intelligence | 18.19 | NatInt Score | 33 |
| Open LLM Leaderboard - GPQA | 3.24 | Score | 29.4 |
| Open LLM Leaderboard - MuSR | 5.93 | Score | 28.6 |
| UGI - Willingness (W/10) | 3.2 | W/10 Score | 24.6 |
| UGI - Writing | 24.96 | Writing Score | 21.9 |
| UGI Leaderboard | 20.43 | UGI Score | 12.2 |
Interactive version: theaggregate.ai/model?slug=configurable-llama-3-1-8b-instruct · How It Works · Data refreshed daily, snapshot 2026-09-05.