NeuralBeagle14-7B — benchmark results

mlabonne's DPO fine-tune of his Beagle14-7B Mistral merge, briefly the top-ranked 7B on the Open LLM Leaderboard in January 2024. Provider: Other. Released 2024-01-15. Access: Open.

Unified ELO 1448 ± 23, rank #1006 of 1776 rated models, from 36 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
EuroEval German NLU - Sb10K59.6Sentiment classification Score (%)93.1
EuroEval Dutch NLU - CoNLL NL63.53Named entity recognition Score (%)73.7
EuroEval German NLU - GermEval64.81Named entity recognition Score (%)73
Open LLM Leaderboard - MuSR12.89Score70.7
EuroEval English Common Sense Reasoning71.96Common Sense Reasoning Average Score (%)69.7
EuroEval Swedish NLU - SUC361.25Named entity recognition Score (%)62.7
EuroEval Swedish NLU - Swerec76.03Sentiment classification Score (%)60.4
EuroEval Dutch Common Sense Reasoning47.87Common Sense Reasoning Average Score (%)59.4
Open LLM Leaderboard - IFEval49.35Score57.8
EuroEval Icelandic NLU - MIM-GOLD NER49.86Named entity recognition Score (%)56.8
EuroEval Swedish Common Sense Reasoning38.78Common Sense Reasoning Average Score (%)54.9
EuroEval German Common Sense Reasoning49.13Common Sense Reasoning Average Score (%)54.7

Interactive version: theaggregate.ai/model?slug=neuralbeagle14-7b · How the rankings work · Data refreshed daily, snapshot 2026-07-22.