diff options
| author | Laurent Mazare <laurent.mazare@gmail.com> | 2024-04-29 09:21:07 +0200 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-04-29 09:21:07 +0200 |
| commit | ed7b99f525ab898aa677fe1f4446e345ac74f4ec (patch) | |
| tree | f614cce52f918a6a89b19615b1a65cdd4eddb2c2 /candle-metal-kernels | |
| parent | 287013ef2864294c3160590150b17e9ca25780af (diff) | |
| download | candle-ed7b99f525ab898aa677fe1f4446e345ac74f4ec.tar.gz candle-ed7b99f525ab898aa677fe1f4446e345ac74f4ec.tar.bz2 candle-ed7b99f525ab898aa677fe1f4446e345ac74f4ec.zip | |
Add a toggle for F16/BF16 accumulation in gemm. (#2141)
* Add a toggle to control f16/bf16 gemm precision.
* Use the faster variant in the quantized example.
* Bugfix.
Diffstat (limited to 'candle-metal-kernels')
0 files changed, 0 insertions, 0 deletions