Llama 3 8B Instruct Ortho Baukit Toxic N128 V3: benchmark results

Provider: Meta.

Aggregate Rating 1415 ± 36, rank #5426 of 10916 rated models, from 6 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open LLM Leaderboard v1 - MMLU66.87Accuracy (%) (5-shot)88.4
Open LLM Leaderboard v1 - GSM8K68.54Accuracy (%) (5-shot)86.2
Open LLM Leaderboard v1 - TruthfulQA MC250.8MC2 (%) (0-shot)49.5
Open LLM Leaderboard v1 - ARC Challenge60.07Normalized accuracy (%) (25-shot)46.2
Open LLM Leaderboard v1 - WinoGrande74.51Accuracy (%) (5-shot)37
Open LLM Leaderboard v1 - HellaSwag78.4Normalized accuracy (%) (10-shot)33.8

Interactive version: theaggregate.ai/model?slug=llama-3-8b-instruct-ortho-baukit-toxic-n128-v3 · How It Works · Data refreshed daily, snapshot 2026-11-08.