salamandra-7B-instruct — benchmark results

Provider: BSC. Released 2024-09-30. Access: Open.

Unified ELO 1352 ± 11, rank #1402 of 1776 rated models, from 285 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
LA Leaderboard - XNLI Galician50.23Accuracy (%)98.5
LA Leaderboard - PIQA Catalan69.97Accuracy (%)95.6
European LLM Leaderboard - Zero-Shot Accuracy54.36Average Accuracy (%)90.2
LA Leaderboard - EusExams Basque40.68Accuracy (%)89.4
LA Leaderboard57.15Average Score (%)88.2
EuroEval Slovak NLU - CSFD Sentiment SK58.48Sentiment classification Score (%)85.9
EuroEval Czech NLU - CSFD Sentiment60.15Sentiment classification Score (%)84.2
LA Leaderboard - Spanish Law Exams38.66Accuracy (%)83.1
LA Leaderboard - AQuAS68.26Accuracy (%)80.9
EuroEval Greek NLU - Greek SA76.02Sentiment classification Score (%)76.4
EuroEval Finnish NLU - Scandisent FI91.2Sentiment classification Score (%)75.3
EuroEval Hungarian NLU - HuSST52.81Sentiment classification Score (%)73.5

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