llama.cpp
imatrix : offload to GPU support
#4957
Merged
Overview
Commits
10
Changes
View On GitHub
Commits
  • backend : add eval callback
    ggerganov committed 2 years ago
  • backend : group nodes in a single compute when user don't need them
    ggerganov committed 2 years ago
  • backend : clean-up the implementation
    ggerganov committed 2 years ago
  • simple : do not perform tensor data copy if not needed
    ggerganov committed 2 years ago
  • simple : fix
    ggerganov committed 2 years ago
  • imatrix : offload to GPU support
    ggerganov committed 2 years ago
  • imatrix : fix ggml_mul_mat_id hanlding
    ggerganov committed 2 years ago
  • ci : add imatrix test
    ggerganov committed 2 years ago
  • ci : rearrange output
    ggerganov committed 2 years ago
  • Merge branch 'master' into gg/imatrix-gpu-4931
    ggerganov committed 2 years ago
FAQ Terms Privacy Refunds Impressum
Loading