mirror of
https://github.com/ggerganov/llama.cpp.git
synced 2024-12-25 02:44:36 +00:00
Bump memory buffer
This commit is contained in:
parent
0c6803321c
commit
7d9ed7b25f
2
main.cpp
2
main.cpp
@ -529,7 +529,7 @@ bool llama_eval(
|
|||||||
|
|
||||||
const int d_key = n_embd/n_head;
|
const int d_key = n_embd/n_head;
|
||||||
|
|
||||||
static size_t buf_size = 256u*1024*1024;
|
static size_t buf_size = 512u*1024*1024;
|
||||||
static void * buf = malloc(buf_size);
|
static void * buf = malloc(buf_size);
|
||||||
|
|
||||||
if (mem_per_token > 0 && mem_per_token*N > buf_size) {
|
if (mem_per_token > 0 && mem_per_token*N > buf_size) {
|
||||||
|
Loading…
Reference in New Issue
Block a user