Johannes Gäßler
|
c63bb1d16a
CUDA: use mul_mat_q kernels by default (#2683)
|
2 жил өмнө |
slaren
|
1123f7fbdf
ggml-cuda : use graph allocator (#2684)
|
2 жил өмнө |
Georgi Gerganov
|
6381d4e110
gguf : new file format with flexible meta data (beta) (#2398)
|
2 жил өмнө |