tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-06-29 20:45:04 +00:00
Code Issues Packages Projects Releases Wiki Activity

llama : update llama_kv_self API

Browse Source 
ggml-ci
This commit is contained in:
Georgi Gerganov
2025-01-14 16:47:34 +02:00
parent fd05ab87aa
commit 17b363afd3
30 changed files with 387 additions and 205 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
examples/llama-bench/llama-bench.cpp
View File

@ -1575,11 +1575,9 @@ int main(int argc, char ** argv) {
return 1;
}
llama_kv_cache * kv = llama_get_kv_cache(ctx);
test t(inst, lmodel, ctx);
llama_kv_cache_clear(kv);
llama_kv_self_clear(ctx);
// cool off before the test
if (params.delay) {
@ -1619,7 +1617,7 @@ int main(int argc, char ** argv) {
}
for (int i = 0; i < params.reps; i++) {
llama_kv_cache_clear(kv);
llama_kv_self_clear(ctx);
uint64_t t_start = get_time_ns();