Llama2.rs: One-file Rust implementation of Llama2

Re:Llama2.rs: One-file Rust implementation of Llama2 by WiSaGaN on Saturday August 05, 2023 @06:12PM (#37016766)

This has 2 dependencies. Notably it depends on rayon.

[ Parent | on HN ]

Re:Llama2.rs: One-file Rust implementation of Llama2 by cchance on Sunday August 06, 2023 @12:20AM (#37018950)

is that bad, a lot of high-performance apps rely on rayon don't they

[ Parent | on HN ]

Re:Llama2.rs: One-file Rust implementation of Llama2 by pests on Sunday August 06, 2023 @01:50AM (#37019335)

A lot of these single-file LLM scripts like the original llama.cpp and the llama2.cpp posted here a bit ago are both self-contained.

[ Parent | on HN ]

Re:Llama2.rs: One-file Rust implementation of Llama2 by dietr1ch on Sunday August 06, 2023 @02:56AM (#37019594)

I think that with C++ projects the only goal there is to keep the build simple since building C++ sucks. This is also the reason why there's many header-only libraries.

StackOverflow| Benefits of header-only libraries: https://stackoverflow.com/questions/12671383/benefits-of-hea...

[ Parent | on HN ]