SmolLM2-135M-Instruct: benchmark results
Provider: HuggingFace. Released 2024-10-31. Access: Open.
Aggregate Rating 1174 ± 23, rank #9750 of 10916 rated models, from 322 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EuroEval Albanian Common Sense Reasoning | 1.99 | Common Sense Reasoning Average Score (%) | 36.8 |
| EuroEval Spanish Summarization - Mlsum ES | 22.56 | chrF3++ (%) | 35.4 |
| Open LLM Leaderboard - MuSR | 38.71 | Score | 33.4 |
| EuroEval Serbian Summarization - LR SUM SR | 19.45 | chrF3++ (%) | 32.7 |
| EuroEval Norwegian Knowledge - Idioms NO | 0.68 | MCC (x100) | 32.4 |
| EuroEval Albanian NLU - ScaLA SQ | 0.92 | Matthews correlation coefficient (%) | 29.5 |
| Open LLM Leaderboard - IFEval | 28.83 | Score | 26 |
| EuroEval Icelandic NLU - ScaLA IS | 0.16 | Matthews correlation coefficient (%) | 25.8 |
| EuroEval Ukrainian Common Sense Reasoning | 1.7 | Common Sense Reasoning Average Score (%) | 24.5 |
| EuroEval Icelandic Knowledge | 1.68 | Knowledge Average Score (%) | 24.4 |
| EuroEval Latvian NLU - ScaLA LV | 0.71 | Matthews correlation coefficient (%) | 24.1 |
| EuroEval Icelandic Common Sense Reasoning | 1.08 | Common Sense Reasoning Average Score (%) | 23.8 |
Interactive version: theaggregate.ai/model?slug=smollm2-135m-instruct · How It Works · Data refreshed daily, snapshot 2026-11-08.