diff options
author | Iwan Kawrakow <iwan.kawrakow@gmail.com> | 2024-08-07 19:24:09 +0300 |
---|---|---|
committer | Kawrakow <48489457+ikawrakow@users.noreply.github.com> | 2024-08-09 16:00:31 +0200 |
commit | 050bdfa101be5b78c2dc2286bad915e2eae21645 (patch) | |
tree | f8f18f7e0bee7fcb4d8c22174e264f851e9b605d /examples/lookup/lookup-create.cpp | |
parent | c3f5e4d9a7ddad8e7af6dd43807815496acddab3 (diff) |
iq6_k: CUDA dot product
90.2 t/s for LLaMA-3.1-8B. Q6_K gives 91.2 t/s, so we are good.
Diffstat (limited to 'examples/lookup/lookup-create.cpp')
0 files changed, 0 insertions, 0 deletions