L3-8B-Lunaris-v1 — benchmark results

Sao10K's generalist-plus-roleplay TIES merge of five Llama 3 8B models, positioned as a more balanced successor to Stheno v3.2. Provider: Other. Released 2024-06-26. Access: Open.

Unified ELO 1465 ± 31, rank #933 of 1776 rated models, from 10 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open LLM Leaderboard - IFEval68.95Score81.8
Open LLM Leaderboard - MMLU-Pro30.97Score64.8
Open LLM Leaderboard - BBH32.11Score60.1
Open LLM Leaderboard - GPQA6.82Score57.4
UGI - Writing32.63Writing Score47.3
Open LLM Leaderboard - MATH Level 59.06Score44.4
UGI - Willingness (W/10)4W/10 Score29.9
Open LLM Leaderboard - MuSR5.55Score26.9
UGI - Natural Intelligence14.57NatInt Score16.9
UGI Leaderboard21.63UGI Score14.5

Interactive version: theaggregate.ai/model?slug=l3-8b-lunaris-v1 · How the rankings work · Data refreshed daily, snapshot 2026-07-22.