SmolLM2-1.7B-Instruct: benchmark results
SmolLM2 1.7B instruction-tuned checkpoint. Provider: HuggingFace. Released 2024-10-31. Access: Open.
Unified ELO 1332 ± 1, rank #1305 of 1392 rated models, from 148 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open CoT - LSAT Analytical Reasoning | 6.09 | CoT Gain (%) | 76.7 |
| Open LLM Leaderboard - IFEval | 53.68 | Score | 64.3 |
| SLM-RAG Arena | 1493.2 | Elo Rating | 56.2 |
| AI Energy Score (Text Generation) | 4 | Energy Score (1-5) | 47.5 |
| LA Leaderboard - AQuAS | 64.81 | Accuracy (%) | 47.1 |
| Open Japanese LLM - CG | 2.01 | Score (%) | 39.2 |
| EuroEval Portuguese NLU - ScaLA PT | 7.15 | Linguistic acceptability Score (%) | 38.6 |
| Open Japanese LLM - Mbpp Pylint Check | 6.43 | Score (%) | 36.1 |
| Open Japanese LLM - Xlsum JA Rouge1 | 22.95 | Score (%) | 35.6 |
| EuroEval Italian NLU - ScaLA IT | 9.31 | Linguistic acceptability Score (%) | 34.9 |
| Open LLM Leaderboard - GPQA | 3.91 | Score | 34.3 |
| EuroEval Spanish NLU - ScaLA ES | 2.54 | Linguistic acceptability Score (%) | 33.6 |
Interactive version: theaggregate.ai/model?slug=smollm2-1-7b-instruct · How It Works · Data refreshed daily, snapshot 2026-09-05.