Gemma 3 1B — benchmark results

Instruction-tuned Gemma 3 1B model row. Provider: Google. Released 2025-03-12. Access: Open.

Unified ELO 1286 ± 19, rank #1562 of 1776 rated models, from 106 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
CLEAR3Version (self-reported)53.6
EuroEval Icelandic NLU - ScaLA IS2.33Linguistic acceptability Score (%)46.1
EuroEval Faroese NLU - ScaLA FO1.99Linguistic acceptability Score (%)43.2
EuroEval Danish NLU - Angry Tweets40.29Sentiment classification Score (%)39.1
EuroEval Swedish NLU - Swerec69.66Sentiment classification Score (%)35.4
EuroEval Norwegian NLU - ScaLA NN7.14Linguistic acceptability Score (%)35.1
EuroEval Norwegian NLU - Norec33.32Sentiment classification Score (%)33.8
EuroEval Faroese NLU - FoQA26.03Reading comprehension Score (%)33.7
EuroEval Italian NLU - SQuAD IT50.73Reading comprehension Score (%)33.3
EuroEval Italian NLU - ScaLA IT7.39Linguistic acceptability Score (%)33.1
EuroEval Danish NLU - ScaLA DA9.58Linguistic acceptability Score (%)32.9
EuroEval Icelandic NLU - NQII32.1Reading comprehension Score (%)32.8

Interactive version: theaggregate.ai/model?slug=gemma-3-1b · How the rankings work · Data refreshed daily, snapshot 2026-07-22.