Skip to content

arg : add --no-mmproj-offload #13093

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 3 commits into from
Apr 24, 2025
Merged

arg : add --no-mmproj-offload #13093

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
b39c925
arg : add --no-mmproj-offload
ngxson Apr 24, 2025
acde5ef
Merge branch 'master' into xsn/clip_no_mmproj_offload
ngxson Apr 24, 2025
c1a49b7
Update common/arg.cpp
ngxson Apr 24, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 7 additions & 0 deletions common/arg.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2140,6 +2140,13 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
params.no_mmproj = true;
}
).set_examples(mmproj_examples));
add_opt(common_arg(
{"--no-mmproj-offload"},
"do not offload multimodal projector to GPU",
[](common_params & params) {
params.mmproj_use_gpu = false;
}
).set_examples(mmproj_examples));
add_opt(common_arg(
{"--image"}, "FILE",
"path to an image file. use with multimodal models. Specify multiple times for batching",
Expand Down
1 change: 1 addition & 0 deletions common/common.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -342,6 +342,7 @@ struct common_params {

// multimodal models (see examples/llava)
struct common_params_model mmproj;
bool mmproj_use_gpu = true; // use GPU for multimodal model
bool no_mmproj = false; // explicitly disable multimodal model
std::vector<std::string> image; // path to image file(s)

Expand Down
7 changes: 4 additions & 3 deletions examples/llava/mtmd-cli.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -40,7 +40,8 @@ static void show_additional_info(int /*argc*/, char ** argv) {
"Usage: %s [options] -m <model> --mmproj <mmproj> --image <image> -p <prompt>\n\n"
" -m and --mmproj are required\n"
" -hf user/repo can replace both -m and --mmproj in most cases\n"
" --image and -p are optional, if NOT provided, the CLI will run in chat mode\n",
" --image and -p are optional, if NOT provided, the CLI will run in chat mode\n"
" to disable using GPU for mmproj model, add --no-mmproj-offload\n",
argv[0]
);
}
Expand Down Expand Up @@ -112,10 +113,10 @@ struct mtmd_cli_context {
void init_vision_context(common_params & params) {
const char * clip_path = params.mmproj.path.c_str();
ctx_vision.reset(mtmd_init_from_file(clip_path, model, mtmd_context_params{
/* use_gpu */ true,
/* use_gpu */ params.mmproj_use_gpu,
/* timings */ true,
/* n_threads */ params.cpuparams.n_threads,
/* verbosity */ GGML_LOG_LEVEL_INFO,
/* verbosity */ params.verbosity > 0 ? GGML_LOG_LEVEL_DEBUG : GGML_LOG_LEVEL_INFO,
}));
if (!ctx_vision.get()) {
LOG_ERR("Failed to load vision model from %s\n", clip_path);
Expand Down
Loading