Skip to content
Open
Show file tree
Hide file tree
Changes from 1 commit
Commits
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
8 changes: 7 additions & 1 deletion modules/llama_cpp_plugin/include/plugin.hpp
Original file line number Diff line number Diff line change
Expand Up @@ -36,10 +36,16 @@ class LlamaCppPlugin : public IPlugin {
virtual std::shared_ptr<ov::ICompiledModel> import_model(std::istream& model,
const ov::SoPtr<ov::IRemoteContext>& context,
const ov::AnyMap& properties) const override;

virtual ov::SupportedOpsMap query_model(const std::shared_ptr<const ov::Model>& model,
const ov::AnyMap& properties) const override;


virtual std::shared_ptr<ov::ICompiledModel> import_model(const ov::Tensor& model, const ov::AnyMap& properties) const override;

virtual std::shared_ptr<ov::ICompiledModel> import_model(const ov::Tensor& model,
const ov::SoPtr<ov::IRemoteContext>& context,
const ov::AnyMap& properties) const override;

private:
size_t m_num_threads = 0;
};
Expand Down
12 changes: 12 additions & 0 deletions modules/llama_cpp_plugin/src/plugin.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -33,6 +33,18 @@ std::shared_ptr<ov::ICompiledModel> LlamaCppPlugin::compile_model(const std::sha
OPENVINO_THROW_NOT_IMPLEMENTED("Currently only direct GGUF file loading is "
"supported for the LLAMA_CPP* plugins");
}

std::shared_ptr<ov::ICompiledModel> LlamaCppPlugin::import_model(const ov::Tensor& model, const ov::AnyMap& properties) const override {
OPENVINO_THROW("This method may not be used with LLAMA_CPP* plugins");
}

std::shared_ptr<ov::ICompiledModel> LlamaCppPlugin::import_model(const ov::Tensor& model,
const ov::SoPtr<ov::IRemoteContext>& context,
const ov::AnyMap& properties) const override {
OPENVINO_THROW("This method may not be used with LLAMA_CPP* plugins");
}


std::shared_ptr<ov::ICompiledModel> LlamaCppPlugin::compile_model(const std::string& fname,
const ov::AnyMap& properties) const {
size_t num_threads = 0;
Expand Down
Loading