SmolLM3-3B-Base: benchmark results

Provider: HuggingFace. Released 2025-07-08. Access: Open.

Aggregate Rating 1391 ± 20, rank #6499 of 10524 rated models, from 356 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
LA Leaderboard - MGSM Spanish15.6Exact Match (%)96.9
LA Leaderboard - MGSM Catalan13.6Exact Match (%)95.4
LA Leaderboard - MGSM Galician13.6Exact Match (%)93.8
LA Leaderboard - OpenBookQA Spanish36.2Accuracy (%)91.2
LA Leaderboard - TELEIA57.14Accuracy (%)91.2
EuroEval German Summarization - Mlsum DE37.45chrF3++ (%)90.5
LA Leaderboard - Summarization Galician6.88BLEU89.6
LA Leaderboard - RagQuAS80.6Semantic Answer Similarity (x100)88.2
LA Leaderboard - xStoryCloze Spanish70.09Accuracy (%)85.3
LA Leaderboard - OpenBookQA Galician28.4Accuracy (%)83.8
EuroEval German NLU - Germanquad61.5Character F1 (%)82.8
LA Leaderboard - caBREU Catalan15.54BLEU82.4

Interactive version: theaggregate.ai/model?slug=smollm3-3b-base · How It Works · Data refreshed daily, snapshot 2026-11-22.