Skip to content
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

Add code path for LLAMA_CPP plugins to load models directly from file #23432

Merged
merged 3 commits into from
Mar 14, 2024
Merged

Add code path for LLAMA_CPP plugins to load models directly from file #23432

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
cc9ae9e
Add code path for LLAMA_CPP plugins to load models directly from file
vshampor Mar 13, 2024
234d72f
Merge the last two ifs instead
vshampor Mar 13, 2024
36bc65a
Fix code style
vshampor Mar 13, 2024
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
10 changes: 3 additions & 7 deletions src/inference/src/dev/core_impl.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -786,8 +786,9 @@ ov::SoPtr<ov::ICompiledModel> ov::CoreImpl::compile_model(const std::string& mod
ov::SoPtr<ov::ICompiledModel> compiled_model;

auto cacheManager = coreConfig.get_cache_config_for_device(plugin, parsed._config)._cacheManager;
// Skip caching for proxy plugin. HW plugin will load network from the cache

if (cacheManager && device_supports_model_caching(plugin) && !is_proxy_device(plugin)) {
// Skip caching for proxy plugin. HW plugin will load network from the cache
CacheContent cacheContent{cacheManager, model_path};
cacheContent.blobId = ov::ModelCache::compute_hash(model_path, create_compile_config(plugin, parsed._config));
std::unique_ptr<CacheGuardEntry> lock = cacheGuard.get_hash_lock(cacheContent.blobId);
Expand All @@ -796,13 +797,8 @@ ov::SoPtr<ov::ICompiledModel> ov::CoreImpl::compile_model(const std::string& mod
auto model = read_model(model_path, std::string{});
return compile_model_and_cache(plugin, model, parsed._config, {}, cacheContent);
});
} else if (cacheManager) {
// this code path is enabled for AUTO / MULTI / BATCH / PROXY devices which don't support
// import / export explicitly, but can redirect this functionality to actual HW plugin
compiled_model = plugin.compile_model(model_path, parsed._config);
} else {
auto model = read_model(model_path, std::string());
compiled_model = plugin.compile_model(model, parsed._config);
compiled_model = plugin.compile_model(model_path, parsed._config);
}
return compiled_model;
}
Expand Down
Loading