Documentation updates #3012

dtrawins · 2025-02-03T08:06:15Z

🛠 Summary

Documentation updates
Staging version http://openvino-doc.iotg.sclab.intel.com/ovms/index.html

🧪 Checklist

Unit tests added.
The documentation updated.
Change follows security best practices.
``

mzegla · 2025-02-03T10:20:12Z

README.md


 ![OVMS picture](docs/ovms_high_level.png)

 The models used by the server need to be stored locally or hosted remotely by object storage services. For more details, refer to [Preparing Model Repository](docs/models_repository.md) documentation. Model server works inside [Docker containers](docs/deploying_server.md#deploying-model-server-in-docker-container), on [Bare Metal](docs/deploying_server.md#deploying-model-server-on-baremetal-without-container), and in [Kubernetes environment](docs/deploying_server.md#deploying-model-server-in-kubernetes).
-Start using OpenVINO Model Server with a fast-forward serving example from the [Quickstart guide](docs/ovms_quickstart.md) or explore [Model Server features](docs/features.md).


Do we want to drop reference to complete feature list in our main readme?

it is included below the key features.

mzegla · 2025-02-03T10:22:42Z

README.md


-* [gRPC](https://grpc.io/)
+* [Simplified Deployments with OpenVINO™ Model Server and TensorFlow Serving](https://community.intel.com/t5/Blogs/Tech-Innovation/Artificial-Intelligence-AI/Simplified-Deployments-with-OpenVINO-Model-Server-and-TensorFlow/post/1353218)


I suppose this blog introduces ovmsclient that is no longer recommended way to prepare a client. Perhaps we should drop this reference?

mzegla · 2025-02-03T10:25:05Z

docs/build_from_source.md

 1. make
 1. bash

+ > **Note**: Building Windows server is covered in [Developer Guide for Windows](windows_developer_guide.md).


Suggested change

> **Note**: Building Windows server is covered in [Developer Guide for Windows](windows_developer_guide.md).

> **Note**: Building on Windows server is covered in [Developer Guide for Windows](windows_developer_guide.md).

mzegla · 2025-02-03T10:25:25Z

docs/build_from_source.md

-Read more detailed usage in [developer guide](https://github.com/openvinotoolkit/model_server/blob/main/docs/developer_guide.md).
-
-## Building ovms.exe on Windows
+Read more detailed about building and testing changes in [developer guide](https://github.com/openvinotoolkit/model_server/blob/main/docs/developer_guide.md).


Suggested change

Read more detailed about building and testing changes in [developer guide](https://github.com/openvinotoolkit/model_server/blob/main/docs/developer_guide.md).

Read more details about building and testing changes in [developer guide](https://github.com/openvinotoolkit/model_server/blob/main/docs/developer_guide.md).

Can we use relative path? This file is not visible to sphinx anyway.

mzegla · 2025-02-03T10:36:28Z

docs/genai.md

@@ -0,0 +1,46 @@
+# Endpoints for Generative Use Cases {#ovms_docs_genai}
+
+OpenVINO Model Server allow extending the REST API interface to support arbitrary input format and execute arbitrary pipeline implemented as a MediaPipe graph.


Suggested change

OpenVINO Model Server allow extending the REST API interface to support arbitrary input format and execute arbitrary pipeline implemented as a MediaPipe graph.

OpenVINO Model Server allows extending the REST API interface to support arbitrary input format and execute arbitrary pipeline implemented as a MediaPipe graph.

mzegla · 2025-02-03T10:40:35Z

docs/genai.md

+It supports a wide range of text generation models from Hugging Faces Hub.
+Internally it employs continuous batching and paged attention algorithms for efficient execution both on CPU and GPU.
+
+Learn model about the [LLM graph configuration](./llm/reference.md) and [exporting the models from Hugging Faces for serving](../demos/common/export_models/README.md).


Suggested change

Learn model about the [LLM graph configuration](./llm/reference.md) and [exporting the models from Hugging Faces for serving](../demos/common/export_models/README.md).

Learn more about the [LLM graph configuration](./llm/reference.md) and [exporting the models from Hugging Face for serving](../demos/common/export_models/README.md).

mzegla · 2025-02-03T10:41:13Z

docs/genai.md

+
+## Text embeddings
+
+Text embeddings transforming the semantic meaning of the text into a numerical vector. This operation is crucial for text searching and algorithms like RAG (Retrieval Augmented Generation).


Suggested change

Text embeddings transforming the semantic meaning of the text into a numerical vector. This operation is crucial for text searching and algorithms like RAG (Retrieval Augmented Generation).

Text embeddings transform the semantic meaning of the text into a numerical vector. This operation is crucial for text searching and algorithms like RAG (Retrieval Augmented Generation).

mzegla · 2025-02-03T10:42:07Z

docs/genai.md

+
+## Documents reranking
+
+Reranking process is used to sort the list of documented based on relevance in the context of a query. Just like text generation and embeddings, it is essential element or RAG chains.


Suggested change

Reranking process is used to sort the list of documented based on relevance in the context of a query. Just like text generation and embeddings, it is essential element or RAG chains.

Reranking process is used to sort the list of documents based on relevance in the context of a query. Just like text generation and embeddings, it is essential element or RAG chains.

mzegla · 2025-02-03T10:44:14Z

docs/starting_server.md

@@ -26,6 +17,10 @@ Start the model server by running the following command with your parameters:
 docker run -d --rm -v <models_repository>:/models -p 9000:9000 -p 8000:8000 openvino/model_server:latest \
 --model_path <path_to_model> --model_name <model_name> --port 9000 --rest_port 8000 --log_level DEBUG
 ```
+or for binary package:
+```c


Is this marking intentional? Or should it be console ?

mzegla · 2025-02-03T10:48:47Z

docs/genai.md

@@ -0,0 +1,46 @@
+# Endpoints for Generative Use Cases {#ovms_docs_genai}


This title is descriptive, but from the menu point of view it's quite long (at the first sight you don't see what it means).
Maybe we could shorten it and start with GenAI phrase so it catches the eye of the viewer like:
GenAI Endpoints or GenAI Use Cases ?

docs/genai.md

Co-authored-by: Damian Kalinowski <[email protected]>

* structure adjustments * deepseek demo

* Documentation updates (#3012) * structure adjustments * deepseek demo * set folder context for demo flow (#3011) Co-authored-by: Dariusz Trawinski <[email protected]> --------- Co-authored-by: Trawinski, Dariusz <[email protected]> Co-authored-by: Dariusz Trawinski <[email protected]>

dtrawins added 2 commits February 2, 2025 01:22

structure adjustments

de79211

improvements

710998a

dtrawins requested review from mzegla and rasapala February 3, 2025 08:06

mzegla reviewed Feb 3, 2025

View reviewed changes

review changes and deepseek demo

2477ed4

mzegla approved these changes Feb 4, 2025

View reviewed changes

link

ce17a1d

dtrawins requested a review from dkalinowski February 4, 2025 11:16

dkalinowski reviewed Feb 4, 2025

View reviewed changes

docs/genai.md Outdated Show resolved Hide resolved

dkalinowski approved these changes Feb 4, 2025

View reviewed changes

Update docs/genai.md

d48c4c6

Co-authored-by: Damian Kalinowski <[email protected]>

dtrawins merged commit f0814fb into releases/2025/0 Feb 4, 2025

dkalinowski pushed a commit that referenced this pull request Feb 6, 2025

Documentation updates (#3012)

1c44a0d

* structure adjustments * deepseek demo

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Documentation updates #3012

Documentation updates #3012

dtrawins commented Feb 3, 2025 •

edited

Loading

mzegla Feb 3, 2025

dtrawins Feb 4, 2025

mzegla Feb 3, 2025

mzegla Feb 3, 2025

mzegla Feb 3, 2025

mzegla Feb 3, 2025

mzegla Feb 3, 2025

mzegla Feb 3, 2025

mzegla Feb 3, 2025

mzegla Feb 3, 2025

mzegla Feb 3, 2025

mzegla Feb 3, 2025


		* [gRPC](https://grpc.io/)
		* [Simplified Deployments with OpenVINO™ Model Server and TensorFlow Serving](https://community.intel.com/t5/Blogs/Tech-Innovation/Artificial-Intelligence-AI/Simplified-Deployments-with-OpenVINO-Model-Server-and-TensorFlow/post/1353218)

	> Note: Building Windows server is covered in [Developer Guide for Windows](windows_developer_guide.md).
	> Note: Building on Windows server is covered in [Developer Guide for Windows](windows_developer_guide.md).

	Read more detailed about building and testing changes in [developer guide](https://github.com/openvinotoolkit/model_server/blob/main/docs/developer_guide.md).
	Read more details about building and testing changes in [developer guide](https://github.com/openvinotoolkit/model_server/blob/main/docs/developer_guide.md).

		@@ -0,0 +1,46 @@
		# Endpoints for Generative Use Cases {#ovms_docs_genai}

		OpenVINO Model Server allow extending the REST API interface to support arbitrary input format and execute arbitrary pipeline implemented as a MediaPipe graph.

	OpenVINO Model Server allow extending the REST API interface to support arbitrary input format and execute arbitrary pipeline implemented as a MediaPipe graph.
	OpenVINO Model Server allows extending the REST API interface to support arbitrary input format and execute arbitrary pipeline implemented as a MediaPipe graph.

	Learn model about the [LLM graph configuration](./llm/reference.md) and [exporting the models from Hugging Faces for serving](../demos/common/export_models/README.md).
	Learn more about the [LLM graph configuration](./llm/reference.md) and [exporting the models from Hugging Face for serving](../demos/common/export_models/README.md).


		## Text embeddings

		Text embeddings transforming the semantic meaning of the text into a numerical vector. This operation is crucial for text searching and algorithms like RAG (Retrieval Augmented Generation).

	Text embeddings transforming the semantic meaning of the text into a numerical vector. This operation is crucial for text searching and algorithms like RAG (Retrieval Augmented Generation).
	Text embeddings transform the semantic meaning of the text into a numerical vector. This operation is crucial for text searching and algorithms like RAG (Retrieval Augmented Generation).


		## Documents reranking

		Reranking process is used to sort the list of documented based on relevance in the context of a query. Just like text generation and embeddings, it is essential element or RAG chains.

Documentation updates #3012

Documentation updates #3012

Conversation

dtrawins commented Feb 3, 2025 • edited Loading

🛠 Summary

🧪 Checklist

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

dtrawins commented Feb 3, 2025 •

edited

Loading