SOURCE-REVIEWED MODEL PROFILE
SmolLM3-3B
A 3B instruction model with optional reasoning, six supported languages and long-context training. Hugging Face publishes its training details and data-mixture information.
Metadata reviewed 2026-10-24 · Original creator: HuggingFaceTB
What it does
- Thinking and non-thinking modes
- English, French, Spanish, German, Italian and Portuguese
- 64K trained context; optional YaRN extension
Local fit
CPU / small GPU. Try language practice or a branching mystery. The 128K extension requires compatible YaRN settings and sufficient memory.
Configuration value; backend support and enough memory are required. Start with a smaller context on limited hardware.
Evidence and limits
Source reviewed; no local generation benchmark for this exact checkpoint.
Connect a supported local text/vision backend; not independently tested
License and source
Keep upstream notices and check the exact downloaded artifact and parent-model terms. The original creator remains credited.
Read the original model card ↗
Reviewed source revision: a07cc9a04f16550a088caea529712d1d335b0ac1