Lumimaid-v0.2-8B — benchmark results

NeverSleep's roleplay fine-tune of Llama 3.1 8B Instruct on a heavily cleaned RP dataset, the NSFW-capable successor to Lumimaid 0.1. Provider: Other. Released 2024-07-24. Access: Open.

Unified ELO 1467 ± 23, rank #924 of 1776 rated models, from 10 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open LLM Leaderboard - GPQA8.17Score68.6
Open LLM Leaderboard - MuSR12.32Score67.2
Open LLM Leaderboard - MATH Level 514.35Score60.8
Open LLM Leaderboard - IFEval50.38Score59.6
Open LLM Leaderboard - BBH31.96Score59.3
Open LLM Leaderboard - MMLU-Pro29.29Score56.3
UGI - Willingness (W/10)4.8W/10 Score36.1
UGI - Writing28.33Writing Score31.9
UGI Leaderboard24.5UGI Score18.5
UGI - Natural Intelligence9.71NatInt Score6.5

Interactive version: theaggregate.ai/model?slug=lumimaid-v0-2-8b · How the rankings work · Data refreshed daily, snapshot 2026-07-22.