Llama 3.2 3B Instruct

Conditional

Meta · Llama · Released 2024-09

3BText

A small edge/on-device model that trades raw capability for speed and a tiny memory footprint, fine for summarization and simple assistants.

Strengths

  • +Runs on phones and low-VRAM devices
  • +Fast inference, low latency
  • +128K context window

Limitations

  • -Noticeably weaker reasoning than 8B+ models
  • -Limited world knowledge relative to larger Llama variants

License

Llama 3.2 Community License

free with acceptable-use or scale conditions - read it

Hardware

runs on a good consumer GPU (or Apple Silicon) with quantization

Links

Stats

meta-llama/Llama-3.2-3B-Instruct— downloads·— likes

via Hugging Face

Related models