CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-10-30 08:42:00 +00:00
Code Issues Packages Projects Releases Wiki Activity

server : pass default --keep argument (#14120)

Browse Source
This commit is contained in:
Taylor
2025-06-11 06:43:43 -04:00
committed by GitHub
parent 7ae2932116
commit 2baf07727f
1 changed files with 2 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
tools/server/server.cpp
View File

@@ -233,6 +233,7 @@ struct server_task {
slot_params defaults; slot_params defaults;
defaults.sampling = params_base.sampling; defaults.sampling = params_base.sampling;
defaults.speculative = params_base.speculative; defaults.speculative = params_base.speculative;
defaults.n_keep = params_base.n_keep;
// enabling this will output extra debug information in the HTTP responses from the server // enabling this will output extra debug information in the HTTP responses from the server
params.verbose = params_base.verbosity > 9; params.verbose = params_base.verbosity > 9;
@@ -2060,6 +2061,7 @@ struct server_context {
SLT_INF(slot, "new slot n_ctx_slot = %d\n", slot.n_ctx); SLT_INF(slot, "new slot n_ctx_slot = %d\n", slot.n_ctx);
slot.params.sampling = params_base.sampling; slot.params.sampling = params_base.sampling;
slot.params.n_keep = params_base.n_keep;
slot.callback_on_release = [this](int) { slot.callback_on_release = [this](int) {
queue_tasks.pop_deferred_task(); queue_tasks.pop_deferred_task();