diff options
| author | Johannes Gäßler <johannesg@5d6.de> | 2023-07-29 23:04:44 +0200 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2023-07-29 23:04:44 +0200 |
| commit | 11f3ca06b8c66b0427aab0a472479da22553b472 (patch) | |
| tree | 8e934ff0d93a78447d996b00561f7ff826c3533f /media/llama0-logo.png | |
| parent | 9baf9ef304f330009d5a93b7390280a0fd27c9a1 (diff) | |
CUDA: Quantized matrix matrix multiplication (#2160)
* mmq implementation for non k-quants
* q6_K
* q2_K
* q3_k
* q4_K
* vdr
* q5_K
* faster q8_1 loading
* loop unrolling
* add __restrict__
* q2_K sc_high
* GGML_CUDA_MMQ_Y
* Updated Makefile
* Update Makefile
* DMMV_F16 -> F16
* Updated README, CMakeLists
* Fix CMakeLists.txt
* Fix CMakeLists.txt
* Fix multi GPU out-of-bounds
Diffstat (limited to 'media/llama0-logo.png')
0 files changed, 0 insertions, 0 deletions