Daniël de Kok on Mythryl: “So cool! transformers can now run GGUF models efficiently. To bring…”

So cool! transformers can now run GGUF models efficiently. To bring the performance close to llama.cpp, it uses the underlying ggml kernels through our kernels library. Awesome work by @marcsun.bsky.social et al.! huggingface.co/blog/transfo...

huggingface.coTransformers now runs llama.cpp quantsWe’re on a journey to advance and democratize artificial intelligence through open source and open science.

22/9/26, 12:51 pm0 reposts18 likes

Post