cturan/llama.cpp

mirror da https://github.com/cturan/llama.cpp

Autore	SHA1 Messaggio	Data
Johannes Gäßler	c63bb1d16a CUDA: use mul_mat_q kernels by default (#2683)	2 anni fa
slaren	1123f7fbdf ggml-cuda : use graph allocator (#2684)	2 anni fa
Georgi Gerganov	6381d4e110 gguf : new file format with flexible meta data (beta) (#2398)	2 anni fa