diff --git a/docs/OV_Runtime_UG/img/applying_low_latency.png b/docs/OV_Runtime_UG/img/applying_low_latency.png
deleted file mode 100644
index bad3a9ef90d..00000000000
--- a/docs/OV_Runtime_UG/img/applying_low_latency.png
+++ /dev/null
@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:f4f6e9d35869fa2c414e58914aaec1607eb7d4768b69c0cbcce5d5fa3ceddba3
-size 56444
diff --git a/docs/OV_Runtime_UG/img/applying_low_latency.svg b/docs/OV_Runtime_UG/img/applying_low_latency.svg
new file mode 100644
index 00000000000..e9b396e944b
--- /dev/null
+++ b/docs/OV_Runtime_UG/img/applying_low_latency.svg
@@ -0,0 +1,2437 @@
+
+
diff --git a/docs/OV_Runtime_UG/img/applying_low_latency_2.png b/docs/OV_Runtime_UG/img/applying_low_latency_2.png
deleted file mode 100644
index 8a76a2d5314..00000000000
--- a/docs/OV_Runtime_UG/img/applying_low_latency_2.png
+++ /dev/null
@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:26ff5d3d42b9838a14481425af8fe8aed791b26fc00a062b91128ba9d5528549
-size 743788
diff --git a/docs/OV_Runtime_UG/img/applying_low_latency_2.svg b/docs/OV_Runtime_UG/img/applying_low_latency_2.svg
new file mode 100644
index 00000000000..d655500547b
--- /dev/null
+++ b/docs/OV_Runtime_UG/img/applying_low_latency_2.svg
@@ -0,0 +1,3517 @@
+
+
diff --git a/docs/OV_Runtime_UG/img/llt2_use_const_initializer.png b/docs/OV_Runtime_UG/img/llt2_use_const_initializer.png
deleted file mode 100644
index 7567d7499e8..00000000000
--- a/docs/OV_Runtime_UG/img/llt2_use_const_initializer.png
+++ /dev/null
@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:9858dbc95426c44d8f11a86936f586ebf4f1d0b8c88ba389d9f89c2948f58ea3
-size 62051
diff --git a/docs/OV_Runtime_UG/img/llt2_use_const_initializer.svg b/docs/OV_Runtime_UG/img/llt2_use_const_initializer.svg
new file mode 100644
index 00000000000..067325f6ec3
--- /dev/null
+++ b/docs/OV_Runtime_UG/img/llt2_use_const_initializer.svg
@@ -0,0 +1,746 @@
+
+
diff --git a/docs/OV_Runtime_UG/img/low_latency_limitation_1.png b/docs/OV_Runtime_UG/img/low_latency_limitation_1.png
deleted file mode 100644
index 8aea64eacd7..00000000000
--- a/docs/OV_Runtime_UG/img/low_latency_limitation_1.png
+++ /dev/null
@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:28f4e7ee50785e9c571725942e67c899d08e87af3802f6bea4721c64bfdb2bac
-size 21722
diff --git a/docs/OV_Runtime_UG/img/low_latency_limitation_2.png b/docs/OV_Runtime_UG/img/low_latency_limitation_2.png
deleted file mode 100644
index 3fea1052da8..00000000000
--- a/docs/OV_Runtime_UG/img/low_latency_limitation_2.png
+++ /dev/null
@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:0923af3acfb69dd0b88a5edf097e60c2655828b643d8e328561b13b0196c0850
-size 47997
diff --git a/docs/OV_Runtime_UG/lowlatency2.md b/docs/OV_Runtime_UG/lowlatency2.md
index c8ff7770fb6..7da4d03f39b 100644
--- a/docs/OV_Runtime_UG/lowlatency2.md
+++ b/docs/OV_Runtime_UG/lowlatency2.md
@@ -10,7 +10,7 @@ The LowLatency2 transformation changes the structure of the network containing [
## Example of Applying the Transformation:
-
+
After applying the transformation, the `ReadValue` operations can receive other operations as an input, as shown in the picture above. These inputs should set the initial value for initialization of the `ReadValue` operations. However, such initialization is not supported in the current State API implementation. Input values are ignored and the initial values for the `ReadValue` operations are set to 0 unless otherwise specified by the user via [State API](@ref openvino-state-api).
@@ -49,7 +49,7 @@ After applying the transformation, the `ReadValue` operations can receive other
```cpp
InferenceEngine::lowLatency2(cnnNetwork, false);
```
- 
+ 
**State naming rule**: A name of a state is a concatenation of names: original `TensorIterator` operation, parameter of the body, and additional suffix `variable_` + `id` (0-base indexing, new indexing for each `TensorIterator`). Use these rules to predict the name of the inserted state after the transformation is applied. For example:
@@ -86,7 +86,12 @@ After applying the transformation, the `ReadValue` operations can receive other
The only way to change the number iterations of `TensorIterator`/`Loop` layer is to use the `Reshape` feature. However, networks can be non-reshapable. The most common reason is that the value of shapes is hardcoded in a constant somewhere in the network.
- 
+ @sphinxdirective
+
+ .. image:: _static/images/low_latency_limitation_2.svg
+ :scale: 70 %
+
+ @endsphinxdirective
**Current solution:**
diff --git a/docs/OV_Runtime_UG/lowlatency_deprecated.md b/docs/OV_Runtime_UG/lowlatency_deprecated.md
index 73a5cdf2516..4d27e503c2c 100644
--- a/docs/OV_Runtime_UG/lowlatency_deprecated.md
+++ b/docs/OV_Runtime_UG/lowlatency_deprecated.md
@@ -2,7 +2,7 @@
The deprecated LowLatency transformation changes the structure of the network containing [TensorIterator](../ops/infrastructure/TensorIterator_1.md) and [Loop](../ops/infrastructure/Loop_5.md) operations by adding the ability to work with the state, inserting the [Assign](../ops/infrastructure/Assign_3.md)/[ReadValue](../ops/infrastructure/ReadValue_3.md) layers, as shown in the picture below.
-
+
After applying the transformation, `ReadValue` operations can receive other operations as an input, as shown in the picture above. These inputs should set the initial value for initialization of `ReadValue` operations. However, such initialization is not supported in the current State API implementation. Input values are ignored and the initial values for the `ReadValue` operations are set to 0 unless otherwise specified by the user via [State API](@ref openvino-state-api).
@@ -73,7 +73,12 @@ After applying the transformation, `ReadValue` operations can receive other oper
Unnecessary parameters may remain on the graph after applying the transformation. The automatic handling of this case inside the transformation is currently not possible. Such parameters should be removed manually from `ngraph::Function` or replaced with a constant.
- 
+ @sphinxdirective
+
+ .. image:: _static/images/low_latency_limitation_1.svg
+ :scale: 70 %
+
+ @endsphinxdirective
**Current solutions:**
* Replace a parameter with a constant (freeze) with the `[0, 0, 0 … 0]` value via [ModelOptimizer CLI](../MO_DG/prepare_model/convert_model/Converting_Model.md): the `--input` or `--freeze_placeholder_with_value` parameters.
@@ -99,7 +104,12 @@ After applying the transformation, `ReadValue` operations can receive other oper
Networks can be non-reshapable. The most common reason is that the value of shapes is hardcoded in the constant somewhere in the network.
- 
+ @sphinxdirective
+
+ .. image:: _static/images/low_latency_limitation_2.svg
+ :scale: 70 %
+
+ @endsphinxdirective
**Current solutions:**
* Trim non-reshapable layers via [ModelOptimizer CLI](../MO_DG/prepare_model/convert_model/Converting_Model.md): the `--input` and `--output` parameters. For example, the parameter and the problematic constant (as shown in the picture above) can be trimmed using the `--input Reshape_layer_name` command-line option.
diff --git a/docs/_static/images/low_latency_limitation_1.svg b/docs/_static/images/low_latency_limitation_1.svg
new file mode 100644
index 00000000000..497ef2ba0ff
--- /dev/null
+++ b/docs/_static/images/low_latency_limitation_1.svg
@@ -0,0 +1,1048 @@
+
+
diff --git a/docs/_static/images/low_latency_limitation_2.svg b/docs/_static/images/low_latency_limitation_2.svg
new file mode 100644
index 00000000000..7d46076dda5
--- /dev/null
+++ b/docs/_static/images/low_latency_limitation_2.svg
@@ -0,0 +1,1613 @@
+
+