HyperLlama-3.1-8B — benchmark results

bunnycore's model_stock merge of three Llama 3.1 8B fine-tunes: Replete-LLM V2, Ultra-Instruct, and Lexi Uncensored V2. Provider: Other. Released 2024-09-04. Access: Open.

Unified ELO 1472 ± 48, rank #902 of 1776 rated models, from 7 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open LLM Leaderboard - IFEval78.83Score95.7
Open Korean LLM Leaderboard447.04Average Score (%)82.3
Open LLM Leaderboard - MATH Level 518.28Score68
Open LLM Leaderboard - MMLU-Pro30.92Score64.5
Open LLM Leaderboard - BBH29.81Score51
Open LLM Leaderboard - GPQA4.92Score41.7
Open LLM Leaderboard - MuSR7.93Score37.1

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