SmolLM2-360M: benchmark results
Provider: HuggingFace. Released 2024-10-31. Access: Open.
Aggregate Rating 1194 ± 21, rank #1571 of 1640 rated models, from 324 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EuroEval Dutch Simplification - Duidelijke Taal | 46.13 | Score (%) | 42 |
| EuroEval Spanish Summarization - Mlsum ES | 23.13 | chrF3++ (%) | 40.8 |
| Open LLM Leaderboard - MuSR | 39.54 | Score | 38.5 |
| EuroEval Serbian Summarization - LR SUM SR | 19.2 | chrF3++ (%) | 32 |
| EuroEval Spanish NLU - ScaLA ES | 1.47 | Matthews correlation coefficient (%) | 29.6 |
| EuroEval Romanian Common Sense Reasoning | 3.43 | Common Sense Reasoning Average Score (%) | 29.1 |
| EuroEval Portuguese NLU - ScaLA PT | 2.79 | Matthews correlation coefficient (%) | 28.9 |
| EuroEval Icelandic NLU - ScaLA IS | 0.37 | Matthews correlation coefficient (%) | 28.5 |
| EuroEval Latvian NLU - ScaLA LV | 1.5 | Matthews correlation coefficient (%) | 28.4 |
| EuroEval Albanian Summarization - LR SUM SQ | 21.5 | chrF3++ (%) | 26.8 |
| EuroEval Catalan Summarization - Dacsa CA | 28.66 | chrF3++ (%) | 26 |
| EuroEval English NLU - SQuAD | 61.31 | Character F1 (%) | 25.2 |
Interactive version: theaggregate.ai/model?slug=smollm2-360m · How It Works · Data refreshed daily, snapshot 2026-10-16.