vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2025-09-16 20:09:41 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Merge branch 'upstream' into concedo_experimental

Browse source 
# Conflicts:
#	docs/development/HOWTO-add-model.md
#	docs/development/token_generation_performance_tips.md
#	flake.lock
This commit is contained in:
Concedo 2024-07-16 00:49:34 +08:00
commit e707ab9025
11 changed files with 549 additions and 171 deletions
Download patch file Download diff file Expand all files Collapse all files

5
common/common.cpp
View file

@ -798,6 +798,10 @@ bool gpt_params_find_arg(int argc, char ** argv, const std::string & arg, gpt_pa
params.cont_batching = true;
return true;
}
if (arg == "-nocb" || arg == "--no-cont-batching") {
params.cont_batching = false;
return true;
}
if (arg == "-fa" || arg == "--flash-attn") {
params.flash_attn = true;
return true;
@ -1539,6 +1543,7 @@ void gpt_params_print_usage(int /*argc*/, char ** argv, const gpt_params & param
options.push_back({ "*", "-np, --parallel N", "number of parallel sequences to decode (default: %d)", params.n_parallel });
options.push_back({ "*", "-ns, --sequences N", "number of sequences to decode (default: %d)", params.n_sequences });
options.push_back({ "*", "-cb, --cont-batching", "enable continuous batching (a.k.a dynamic batching) (default: %s)", params.cont_batching ? "enabled" : "disabled" });
options.push_back({ "*", "-nocb, --no-cont-batching", "disable continuous batching" });
options.push_back({ "multi-modality" });
options.push_back({ "*", " --mmproj FILE", "path to a multimodal projector file for LLaVA. see examples/llava/README.md" });