SmolLM2-135M: benchmark results
Provider: HuggingFace. Released 2024-10-31. Access: Open.
Aggregate Rating 1160 ± 26, rank #9833 of 10916 rated models, from 288 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard - MuSR | 41.12 | Score | 51.4 |
| Fibble3 Arena (8 guesses) | 0 | Daily clean-run win rate (%), 8 guesses | 50 |
| Fibble4 Arena (8 guesses) | 0 | Daily clean-run win rate (%), 8 guesses | 46.4 |
| Fibble2 Arena (8 guesses) | 0 | Daily clean-run win rate (%), 8 guesses | 43.8 |
| EuroEval Norwegian Knowledge - Idioms NO | 1.78 | MCC (x100) | 42.5 |
| Fibble2 Arena | 0 | Daily clean-run win rate (%), 10 guesses | 40 |
| Fibble4 Arena | 0 | Daily clean-run win rate (%), 12 guesses | 38.9 |
| Fibble3 Arena | 0 | Daily clean-run win rate (%), 12 guesses | 36.8 |
| Fibble5 Arena | 0 | Daily clean-run win rate (%), 9 guesses | 35.9 |
| EuroEval Catalan NLU - ScaLA CA | 0.01 | Matthews correlation coefficient (%) | 34.6 |
| EuroEval Icelandic NLU - ScaLA IS | 0.87 | Matthews correlation coefficient (%) | 34.3 |
| Fibble5 Arena (8 guesses) | 0 | Daily clean-run win rate (%), 8 guesses | 33.3 |
Interactive version: theaggregate.ai/model?slug=smollm2-135m · How It Works · Data refreshed daily, snapshot 2026-11-08.