Llama 2 13B Base — benchmark results

Provider: Meta. Released 2023-07-18. Access: Open.

Unified ELO 1320 ± 22, rank #1492 of 1776 rated models, from 113 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open Japanese LLM - Xlsum JA Bleu JA75.69Score (%)99.3
EuroEval Portuguese NLU - MultiWikiQA PT76.05Reading comprehension Score (%)88.9
EuroEval Italian NLU - SQuAD IT71.05Reading comprehension Score (%)76.8
EuroEval Dutch NLU - DBRD90.19Sentiment classification Score (%)72.6
EuroEval Spanish NLU - MLQA ES63.41Reading comprehension Score (%)72.4
EuroEval Portuguese NLU - SST-2 PT80.59Sentiment classification Score (%)70.5
Open Japanese LLM - Wiki Dependency SET F136.98Score (%)68
URIAL-Bench - Writing7.53Judge Score (0-10)61.1
OpenEval - BBQ61.66Exact Match (%)59.5
EuroEval Portuguese NLU50.31NLU Average Score (%)52.7
URIAL-Bench - Extraction4.7Judge Score (0-10)50
EuroEval Danish Knowledge46.76Knowledge Average Score (%)47.7

Interactive version: theaggregate.ai/model?slug=llama-2-13b-base · How the rankings work · Data refreshed daily, snapshot 2026-07-22.