Llama 3 8B Ee: benchmark results

Provider: Meta. Access: Open.

Unified ELO 1459 ± 20, rank #1651 of 2928 rated models, from 12 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open LLM Leaderboard v1 - MMLU66.01Accuracy (%) (5-shot)85.7
Open LLM Leaderboard - GPQA31.38Score70.7
Open LLM Leaderboard v1 - GSM8K39.88Accuracy (%) (5-shot)55.1
Open LLM Leaderboard v1 - WinoGrande77.11Accuracy (%) (5-shot)53.4
Open LLM Leaderboard v1 - HellaSwag81.48Normalized accuracy (%) (10-shot)45.2
Open LLM Leaderboard - MMLU-Pro32.2Score44.2
Open LLM Leaderboard v1 - ARC Challenge58.62Normalized accuracy (%) (25-shot)40.4
Open LLM Leaderboard - BBH46.26Score37
Open LLM Leaderboard - MATH Level 54.83Score27.3
Open LLM Leaderboard v1 - TruthfulQA MC242.9MC2 (%) (0-shot)21.9
Open LLM Leaderboard - MuSR36.54Score20.8
Open LLM Leaderboard - IFEval19.51Score11.2

Interactive version: theaggregate.ai/model?slug=llama-3-8b-ee · How It Works · Data refreshed daily, snapshot 2026-09-23.