Back to Freedom.Tech Back
All llama.cpp releasesAll versions
Release Fri, Jul 31, 2026 1 min read

llama.cpp b10212

Original release notes

llama : load MTP tensors only if they are really used (#26296)

  • llama : load MTP tensors only if they are really used
  • llama : skip loading MTP (if not used) in remaining models that support MTP

---------

Co-authored-by: Stanisaw Szymczyk

Website: -

UI: