salamandra-2B-instruct — benchmark results

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

Unified ELO 1271 ± 45, rank #1587 of 1776 rated models, from 19 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
LA Leaderboard - XNLI Galician46.84Accuracy (%)75
LA Leaderboard - PIQA Catalan64.31Accuracy (%)69.1
EuroEval Portuguese NLU - SST-2 PT73.31Sentiment classification Score (%)43
EuroEval Finnish NLU - Scandisent FI85.18Sentiment classification Score (%)33.2
LA Leaderboard - AQuAS62.14Accuracy (%)30.9
EuroEval Icelandic Knowledge2.48Knowledge Average Score (%)28.2
EuroEval Dutch NLU - DBRD64.59Sentiment classification Score (%)24.5
EuroEval Norwegian Knowledge2.94Knowledge Average Score (%)21.2
EuroEval Portuguese NLU - ScaLA PT1.14Linguistic acceptability Score (%)20.7
EuroEval Italian NLU - Sentipolc1629.8Sentiment classification Score (%)20.1
LA Leaderboard48.81Average Score (%)19.1
LA Leaderboard - COPA Spanish69.4Accuracy (%)19.1

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