Skip To Main Content
Support Knowledge Base

How to Use Model Caching on NPU in C++?

Content Type: Product Information & Documentation   |   Article ID: 000101134   |   Last Reviewed: 05/22/2025

Description

Unable to determine how to use model caching on NPU in C++.

Resolution

Specify pipeline_config for NPU pipeline to export and import caching compiled models.

Export example:

ov::AnyMap pipeline_config = { { "EXPORT_BLOB", "YES" }, { "BLOB_PATH", ".npucache\\compiled_model.blob" } };
ov::genai::WhisperPipeline pipeline(model_path, "NPU", pipeline_config);

Import example:

ov::AnyMap pipeline_config = { { "BLOB_PATH", ".npucache\\compiled_model.blob" } };
ov::genai::WhisperPipeline pipeline(model_path, "NPU", pipeline_config);

Related Products

This article applies to 3 products.
Intel® Xeon Phi™ Processor Software OpenVINO™ toolkit Performance Libraries