llama.cpp
The fast engine that many local AI apps use to run language models on everyday hardware
The fast engine that many local AI apps use to run language models on everyday hardware. Built for: Builders who want to run models efficiently on their own hardware. Format: Library / CLI. License: MIT. Open source on GitHub (130,841 stars as of 2026-10-11).
Available on
CLILibrary
Pricing
Free and open source; self-hosting, model or API costs may apply
Free tier availablefree pricing
Price checked Oct 12, 2026
Visit llama.cpp (opens in new tab)Compare with...