Llama-3.1-8B-Lexi-Uncensored-V2: benchmark results
Orenguteng's uncensored fine-tune of Llama 3.1 8B Instruct, tuned for high compliance with guardrails removed. Provider: Orenguteng. Released 2024-08-09. Access: Open.
Aggregate Rating 1444 ± 19, rank #1028 of 1656 rated models, from 20 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-IFEval | 51.11 | Strict accuracy, prompt/instruction mean (%) | 94.3 |
| Open LLM Leaderboard - IFEval | 77.92 | Score | 94 |
| Open Korean LLM - Ko-EQ-Bench | 33.07 | EQ-Bench score | 82.8 |
| Open Korean LLM Leaderboard | 41.37 | Average Score (%) | 82.3 |
| Open Ko-LLM Leaderboard | 41.37 | Average (%) | 82.2 |
| Open Korean LLM - Ko-GPQA-Diamond | 25.76 | Normalized accuracy (%) | 81.3 |
| Open LLM Leaderboard - MATH Level 5 | 19.71 | Score | 70.5 |
| Open LLM Leaderboard - MMLU-Pro | 37.81 | Score | 64.4 |
| UGI - Willingness (W/10) | 6.2 | W/10 Score | 59 |
| Open Korean LLM - KorNAT-CKA | 25.3 | Normalized accuracy (%) | 56.7 |
| Open LLM Leaderboard - BBH | 50.84 | Score | 52.5 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 18.95 | Exact match, flexible extract (%) | 50.3 |
Interactive version: theaggregate.ai/model?slug=llama-3-1-8b-lexi-uncensored-v2 · How It Works · Data refreshed daily, snapshot 2026-10-20.