Merge branch 'master' into support_offline_cpu

docs/articles_en/about-openvino/performance-benchmarks/generative-ai-performance.rst

@@ -17,27 +17,30 @@ The tables below list the key performance indicators for inference on built-in G
 
    .. tab-item:: 9-288V
 
-      .. csv-table::
+      .. data-table::
          :class: modeldata stripe
          :name: supportedModelsTable_V1
          :header-rows: 1
          :file:  ../../_static/benchmarks_files/llm_models_9-288V.csv
+         :hidden: [3,4,6]
 
    .. tab-item:: 7-268V
 
-      .. csv-table::
+      .. data-table::
          :class: modeldata stripe
          :name: supportedModelsTable_V2
          :header-rows: 1
          :file:  ../../_static/benchmarks_files/llm_models_7-258V.csv
+         :hidden: [3,4,6]
 
    .. tab-item:: 7-155H
 
-      .. csv-table::
+      .. data-table::
          :class: modeldata stripe
          :name: supportedModelsTable_V3
          :header-rows: 1
          :file:  ../../_static/benchmarks_files/llm_models_7-155H.csv
+         :hidden: [3,4,6]
 
 
 .. grid:: 1 1 2 2

docs/articles_en/about-openvino/release-notes-openvino.rst

@@ -39,8 +39,9 @@ OpenVINO™ Runtime
 CPU Device Plugin
 -----------------------------------------------------------------------------------------------
 
-* KV cache now uses asymmetric U8 as the default precision, reducing memory stress for LLMs and
-  increasing their performance. This option can be controlled by model meta data.
+* KV cache now uses asymmetric 8-bit unsigned integer (U8) as the default precision, reducing
+  memory stress for LLMs and increasing their performance. This option can be controlled by
+  model meta data.
 * Quality and accuracy has been improved for selected models with several bug fixes.
 
 GPU Device Plugin

docs/articles_en/assets/snippets/ov_caching.cpp

@@ -90,42 +90,14 @@ auto compiled = core.compile_model(model, device, config);            // Step 5:
     }
 }
 
-void part5() {
-    std::string modelPath = "/tmp/myModel.xml";
-    std::string device = "GPU";
-    ov::Core core;                                           // Step 1: create ov::Core object
-    core.set_property(ov::cache_dir("/path/to/cache/dir"));  // Step 1b: Enable caching
-//! [ov:caching:part5]
-static const char codec_key[] = {0x30, 0x60, 0x70, 0x02, 0x04, 0x08, 0x3F, 0x6F, 0x72, 0x74, 0x78, 0x7F};
-auto codec_xor = [&](const std::string& source_str) {
-    auto key_size = sizeof(codec_key);
-    int key_idx = 0;
-    std::string dst_str = source_str;
-    for (char& c : dst_str) {
-        c ^= codec_key[key_idx % key_size];
-        key_idx++;
-    }
-    return dst_str;
-};
-auto compiled = core.compile_model(modelPath,
-                                   device,
-                                   ov::cache_encryption_callbacks(ov::EncryptionCallbacks{codec_xor, codec_xor}),
-                                   ov::cache_mode(ov::CacheMode::OPTIMIZE_SIZE));  // Step 5: Compile model
-//! [ov:caching:part5]
-    if (!compiled) {
-        throw std::runtime_error("error");
-    }
-}
-
 int main() {
     try {
         part0();
         part1();
         part2();
         part3();
         part4();
-        part5();
     } catch (...) {
     }
     return 0;
-}
+}

docs/articles_en/assets/snippets/ov_caching.py

@@ -59,20 +59,3 @@ def decrypt_base64(src):
 model = core.read_model(model=model_path)
 compiled_model = core.compile_model(model=model, device_name=device_name, config=config_cache)
 # ! [ov:caching:part4]
-
-# ! [ov:caching:part5]
-import base64
-
-def encrypt_base64(src):
-    return base64.b64encode(bytes(src, "utf-8"))
-
-def decrypt_base64(src):
-    return base64.b64decode(bytes(src, "utf-8"))
-
-core = ov.Core()
-core.set_property({props.cache_dir: path_to_cache_dir})
-config_cache = {}
-config_cache["CACHE_ENCRYPTION_CALLBACKS"] = [encrypt_base64, decrypt_base64]
-config_cache["CACHE_MODE"] = "OPTIMIZE_SIZE"
-compiled_model = core.compile_model(model=model_path, device_name='GPU', config=config_cache)
-# ! [ov:caching:part5]

docs/articles_en/openvino-workflow/running-inference/optimize-inference/optimizing-latency/model-caching-overview.rst

@@ -139,7 +139,7 @@ To check in advance if a particular device supports model caching, your applicat
 Set "cache_encryption_callbacks" config option to enable cache encryption
 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
 
-If model caching is enabled in the CPU Plugin, the model topology can be encrypted while it is saved to the cache and decrypted when it is loaded from the cache. Currently, this property can be set only in ``compile_model``.
+If model caching is enabled, the model topology can be encrypted when saving to the cache and decrypted when loading from the cache. This property can currently be set only in ``compile_model``.
 
 .. tab-set::
 
@@ -157,24 +157,6 @@ If model caching is enabled in the CPU Plugin, the model topology can be encrypt
          :language: cpp
          :fragment: [ov:caching:part4]
 
-If model caching is enabled in the GPU Plugin, the model topology can be encrypted while it is saved to the cache and decrypted when it is loaded from the cache. Full encryption only works when the ``CacheMode`` property is set to ``OPTIMIZE_SIZE``.
-
-.. tab-set::
-
-   .. tab-item:: Python
-      :sync: py
-
-      .. doxygensnippet:: docs/articles_en/assets/snippets/ov_caching.py
-         :language: py
-         :fragment: [ov:caching:part5]
-
-   .. tab-item:: C++
-      :sync: cpp
-
-      .. doxygensnippet:: docs/articles_en/assets/snippets/ov_caching.cpp
-         :language: cpp
-         :fragment: [ov:caching:part5]
-
 .. important::
 
-   Currently, this property is supported only by the CPU and GPU plugins. For other HW plugins, setting this property will not encrypt/decrypt the model topology in cache and will not affect performance.
+   Currently, this property is supported only by the CPU plugin. For other HW plugins, setting this property will not encrypt/decrypt the model topology in cache and will not affect performance. 

docs/sphinx_setup/_static/js/custom.js

@@ -416,7 +416,7 @@ document.addEventListener('DOMContentLoaded', function () {
             }
 
             await element.initialize({
-                accessToken: "xx1f2aebd3-4307-4632-aeea-17c13378b237",
+                accessToken: "xx2b580d60-addf-451d-94fd-06effafb7686",
                 organizationId: "intelcorporationproductione78n25s6"
             });