vrr/prima.cpp
2
0
Fork 0
mirror of https://github.com/Lizonghang/prima.cpp.git synced 2025-09-10 14:04:34 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

fix compute buffer estimate: tested on metal

Browse source
This commit is contained in:
Li, Zonghang 2025-06-20 13:43:55 +04:00
parent dd589561b4
commit 80e5b71b48
5 changed files with 134 additions and 31 deletions
Download patch file Download diff file Expand all files Collapse all files

3
common/arg.cpp
View file

@ -749,6 +749,9 @@ gpt_params_context gpt_params_parser_init(gpt_params & params, llama_example ex,
format("maximum GPU memory to use (default: %d)", params.gpu_mem),
[](gpt_params & params, int value) {
params.gpu_mem = value; // in GiB
if (value == 0) {
LOG_WRN("WARN: Set --gpu-mem to 0 may lead to errors during workload distribution.\n");
}
}
).set_env("LLAMA_ARG_CUDA_MEM"));
add_opt(llama_arg(