jdelony/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-06-27 23:50:20 -05:00
Code Issues Packages Projects Releases Wiki Activity

server: fix router args not being forwarded to child instances (#24760)

Browse Source
This commit is contained in:
Xuan-Son Nguyen 2026-06-18 12:15:46 +02:00 committed by GitHub
parent 24bba7b98e
commit 968c43891a
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 6 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
tools/server/server-models.cpp
View File

@ -351,6 +351,12 @@ void server_models::load_models() {
source_map[name] = SERVER_MODEL_SOURCE_PRESET;
}
// overlay router's own CLI args on top of every model preset so that
// e.g. `llama-server --temp 0` is honoured by all child processes
for (auto & [name, preset] : final_presets) {
preset.merge(base_preset);
}
auto get_source = [&](const std::string & name) {
return source_map.count(name) ? source_map.at(name) : SERVER_MODEL_SOURCE_PRESET;
};