BaeZel-8B-LINEAR — benchmark results

DreadPoor's linear merge of three of his own Llama 3.1 8B model-stock merges (Heart_Stolen, Aspire, LemonP). Provider: Other. Released 2024-11-08. Access: Open.

Unified ELO 1525 ± 20, rank #704 of 1776 rated models, from 10 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open LLM Leaderboard - IFEval73.78Score88.4
Open LLM Leaderboard - GPQA9.51Score76
Open LLM Leaderboard - MuSR13.34Score73.6
Open LLM Leaderboard - BBH35.54Score71.9
Open LLM Leaderboard - MMLU-Pro31.79Score68.8
Open LLM Leaderboard - MATH Level 518.13Score67.7
UGI - Willingness (W/10)6.2W/10 Score57.7
UGI Leaderboard31.34UGI Score36.6
UGI - Natural Intelligence16.43NatInt Score26.1
UGI - Writing24.81Writing Score21.7

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