SmolLM 1.7B punches above its weight for a model of its size, handling straightforward text tasks like summarization, Q&A, and simple reasoning with surprising competence. It's the compact colleague who gets things done quickly without demanding much compute — though it struggles with complex multi-step reasoning or nuanced tasks that larger models handle more gracefully. Its open weights and permissive license make it easy to inspect, fine-tune, and deploy anywhere.
| Benchmark | Score | Type | Recorded |
|---|---|---|---|
| IFEval | 23.6 | accuracy | 3d ago |