SmolLM2 SmolLM2 is a family of compact language models available in three size: 135M, 360M, and 1.7B parameters. They are capable of solving a wide range of tasks while being lightweight enough to run on-device. You can find our most capable model 🤏 SmolLM2-1.7B-Instruct here. New: Introducing SmolTalk, the SFT dataset of SmolLM2 🚀 Table of Contents Usage Transformers Chat in TRL Local applications Smol-tools Pre-training Fine-tuning Evaluation Synthetic data pipelines Usage Our most powerful ...