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: Meta. Released 2024-08-09. Access: Open.

Unified ELO 1469 ± 32, rank #912 of 1776 rated models, from 11 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open LLM Leaderboard - IFEval77.92Score94
Open Korean LLM Leaderboard449.36Average Score (%)82.6
Open LLM Leaderboard - MATH Level 519.71Score70.5
Open LLM Leaderboard - MMLU-Pro30.9Score64.4
UGI - Willingness (W/10)6.2W/10 Score57.7
Open LLM Leaderboard - BBH29.69Score50.6
Open LLM Leaderboard - GPQA4.36Score37.6
Open LLM Leaderboard - MuSR7.77Score36.2
UGI - Natural Intelligence16.16NatInt Score24.7
UGI Leaderboard25.72UGI Score20.7
UGI - Writing19.4Writing Score12.3

Interactive version: theaggregate.ai/model?slug=llama-3-1-8b-lexi-uncensored-v2 · How the rankings work · Data refreshed daily, snapshot 2026-07-22.