W6 PR2: Llama-3.2-1B on llama.cpp's GGUF (Q8_0/Q4_0/F32): tokens/s vs llama.cpp, perplexity - #42

W6 PR2: Llama-3.2-1B on llama.cpp's GGUF (Q8_0/Q4_0/F32): tokens/s vs llama.cpp, perplexity#42

Mattral merged 4 commits into

mainMattral/simd-ml-kernels:mainfrom

w6/llama-quantMattral/simd-ml-kernels:w6/llama-quantCopy head branch name to clipboard