Compare software

llama.cppFoundry Local
Free

Free and open source under the MIT licence.

 Free

Free to download from Microsoft.

Windows, macOS, Linux, Android, Command line Windows, macOS, Command line
  • No Python and no dependencies; one binary runs the model
  • Quantisation down to 1.5 bits fits big models into small memory
  • Runs models entirely on the device
  • Hardware-accelerated local inference
  • Command line first; the web interface is basic next to dedicated apps
  • Quantising hard reduces answer quality, and the trade-off is yours to judge
  • Oriented toward developers building applications
  • Licence terms are not stated in the research used here
Downloadable app Downloadable app
Open source Closed source
License: MIT License not stated
No account needed Not stated if an account is needed
Works offline Works offline

Checked September 21, 2026

Checked September 23, 2026

Catalog facts only. Anything “not stated” is unconfirmed. Check full listings for details.