SmolLM family

The SmolLM family is a series of small language models (SLMs) developed by hugging-face, optimized for high performance, efficiency, and local deployment.

Models

SmolLM3-3B

A 3-billion parameter model designed for advanced reasoning and local usability.

• Key Features:
  • Reasoning: Features an advanced “thinking mode” that allows the model to display its internal reasoning process prior to generating a final response.
• Deployment:
  • Can be served locally using vLLM (see guide by Fahd Mirza).
• Resources:
  • 2026 04 14 New SmoILM3 from hugging face
  • Hugging Face Blog
  • LLM Tutorials (GitHub)
  • Local Installation Guide (YouTube)