diff --git a/site/docs/getting-started/introduction.mdx b/site/docs/getting-started/introduction.mdx
index 000b4b0bf1..56808c6105 100644
--- a/site/docs/getting-started/introduction.mdx
+++ b/site/docs/getting-started/introduction.mdx
@@ -4,3 +4,58 @@ sidebar_label: Introduction
---
# Introduction to OpenVINO GenAI
+
+## What is OpenVINO GenAI?
+
+OpenVINOβ’ GenAI is a library of the most popular Generative AI model pipelines, optimized execution methods, and samples that run on top of highly performant [OpenVINO Runtime](https://docs.openvino.ai/).
+It provides simplified APIs for running generative models, hiding the complexity of the generation process and enabling developers to easily integrate state-of-the-art generative models into their applications with minimal code.
+
+As a lightweight solution designed for efficient inference, OpenVINO GenAI includes all the core functionality needed for generative model execution (e.g. tokenization via `openvino-tokenizers`) with no external dependencies required.
+This library is friendly to PC and laptop execution, and optimized for resource consumption.
+
+## Key Features and Benefits
+
+- **π¦ Pre-built Generative AI Pipelines:** Ready-to-use pipelines for text generation (LLMs), image generation (Diffuser-based), speech processing (Whisper), and visual language models (VLMs). See all [supported use cases](/docs/category/use-cases).
+- **π£ Minimal Footprint:** Smaller binary size and reduced memory footprint compared to other frameworks.
+- **π Performance Optimization:** Hardware-specific optimizations for CPU, GPU, and NPU devices.
+- **π¨βπ» Programming Language Support:** Comprehensive APIs in both Python and C++.
+- **ποΈ Model Compression:** Support for 8-bit and 4-bit weight compression, including embedding layers.
+- **π Advanced Inference Capabilities:** In-place KV-cache, dynamic quantization, speculative sampling, and more.
+- **π¨ Wide Model Compatibility:** Support for popular models including Llama, Mistral, Phi, Qwen, Stable Diffusion, Flux, Whisper, and others. Refer to the [Supported Models](/docs/supported-models) for more details.
+
+## Workflow Overview
+
+Using OpenVINO GenAI typically involves three main steps:
+
+1. **Model Preparation:**
+ - Convert model from other frameworks to the OpenVINO IR format (e.g. using `optimum-intel`), optionally applying weights compression.
+ - Download pre-converted model in OpenVINO IR format (e.g. from [OpenVINO Toolkit](https://huggingface.co/OpenVINO) organization on Hugging Face).
+ :::info
+
+ You can use models from [Hugging Face](https://huggingface.co/) and [ModelScope](https://modelscope.cn/home)
+
+ :::
+2. **Pipeline Setup:** Initialize the appropriate pipeline for your task (`LLMPipeline`, `Text2ImagePipeline`, `WhisperPipeline`, `VLMPipeline`, etc.) with the converted model.
+3. **Inference:** Run the model with your inputs using the pipeline's simple API.
+
+
+
+## Comparison with Alternatives
+
+Unlike base OpenVINO, which requires manual implementation of generation loops, tokenization, scheduling etc., OpenVINO GenAI provides these components in a ready-to-use package.
+
+Compared to Hugging Face Optimum Intel, OpenVINO GenAI offers a smaller footprint, fewer dependencies, and better performance optimization options, particularly for C++ applications.
+
+| Feature | OpenVINO GenAI | Base OpenVINO | Hugging Face Optimum Intel |
+|-|-|-|-|
+| Easy-to-use APIs | β
| β | β
|
+| Low footprint | β
| β
| β |
+| C++ support | β
| β
| β |
+| Pre-built pipelines | β
| β | β
|
+| Model variety | Medium | High | High |
+
+## System Requirements
+
+OpenVINO GenAI is built on top of OpenVINO Runtime and shares the same system requirements.
+
+Refer to the [OpenVINO System Requirements](https://docs.openvino.ai/2025/about-openvino/release-notes-openvino/system-requirements.html) for more details.
diff --git a/site/docs/supported-models/_category_.json b/site/docs/supported-models/_category_.json
new file mode 100644
index 0000000000..8c0ec25226
--- /dev/null
+++ b/site/docs/supported-models/_category_.json
@@ -0,0 +1,8 @@
+{
+ "label": "Supported Models",
+ "position": 4,
+ "link": {
+ "type": "doc",
+ "id": "supported-models/index"
+ }
+}
diff --git a/site/docs/supported-models/_components/base-models-table/index.tsx b/site/docs/supported-models/_components/base-models-table/index.tsx
new file mode 100644
index 0000000000..e800c461d2
--- /dev/null
+++ b/site/docs/supported-models/_components/base-models-table/index.tsx
@@ -0,0 +1,38 @@
+import Link from '@docusaurus/Link';
+import { Children } from 'react';
+
+type BaseModelsTableProps = {
+ headers: string[];
+ rows: React.JSX.Element[];
+};
+
+export function BaseModelsTable({ headers, rows }: BaseModelsTableProps): React.JSX.Element {
+ return (
+
+
+
+ {headers.map((v) => (
+ {v}
+ ))}
+
+
+ {Children.map(rows, (row) => row)}
+
+ );
+}
+
+export const LinksCell = ({ links }: { links: string[] }) => (
+
+
+ {links.map((link) => (
+
+
+ ))}
+
+
+);
+
+export const StatusCell = ({ value }: { value: boolean }) => (
+ {value ? 'β
' : 'β'}
+);
diff --git a/site/docs/supported-models/_components/image-generation-models-table/index.tsx b/site/docs/supported-models/_components/image-generation-models-table/index.tsx
new file mode 100644
index 0000000000..ad95b52003
--- /dev/null
+++ b/site/docs/supported-models/_components/image-generation-models-table/index.tsx
@@ -0,0 +1,29 @@
+import React from 'react';
+import { BaseModelsTable, LinksCell, StatusCell } from '../base-models-table';
+import { IMAGE_GENERATION_MODELS } from './models';
+
+export default function ImageGenerationModelsTable(): React.JSX.Element {
+ const headers = [
+ 'Architecture',
+ 'Text to Image',
+ 'Image to Image',
+ 'LoRA Support',
+ 'Example HuggingFace Models',
+ ];
+
+ const rows = IMAGE_GENERATION_MODELS.map(
+ ({ architecture, textToImage, imageToImage, loraSupport, links }) => (
+
+
+ {architecture}
+
+
+ )
+ );
+
+ return
+
+ {architecture}
+
+
+ ));
+
+ return
+
+ {architecture}
+
+ {models[0].name}
+
+ {models.slice(1).map(({ name, links }) => (
+
+ {name}
+
+ ))}
+
+ ));
+
+ return
+
+ {architecture}
+
+ {models[0].name}
+
+ {models.slice(1).map(({ name, loraSupport, links }) => (
+
+ {name}
+
+ ))}
+
+ ));
+
+ return
+
+ {architecture}
+
+ {models[0].name}
+
+ {models.slice(1).map(({ name, loraSupport, links }) => (
+
+ {name}
+
+ ))}
+
+ ));
+
+ return
diff --git a/site/docs/use-cases/2-Generating-Images.md b/site/docs/use-cases/2-Generating-Images.md
deleted file mode 100644
index 990cd46747..0000000000
--- a/site/docs/use-cases/2-Generating-Images.md
+++ /dev/null
@@ -1,5 +0,0 @@
----
-sidebar_position: 1
----
-
-# Generating Images Using Diffusers
diff --git a/site/docs/use-cases/2-Image-Generation/_sections/_run_model/_code_example_cpp.mdx b/site/docs/use-cases/2-Image-Generation/_sections/_run_model/_code_example_cpp.mdx
new file mode 100644
index 0000000000..24054cd985
--- /dev/null
+++ b/site/docs/use-cases/2-Image-Generation/_sections/_run_model/_code_example_cpp.mdx
@@ -0,0 +1,16 @@
+import CodeBlock from '@theme/CodeBlock';
+
+
+{`#include "openvino/genai/image_generation/text2image_pipeline.hpp"
+#include "imwrite.hpp"
+
+int main(int argc, char* argv[]) {
+ const std::string models_path = argv[1], prompt = argv[2];
+
+ ov::genai::Text2ImagePipeline pipe(models_path, "${props.device || 'CPU'}");
+ ov::Tensor image = pipe.generate(prompt);
+
+ imwrite("image.bmp", image, true);
+}
+`}
+
diff --git a/site/docs/use-cases/2-Image-Generation/_sections/_run_model/_code_example_python.mdx b/site/docs/use-cases/2-Image-Generation/_sections/_run_model/_code_example_python.mdx
new file mode 100644
index 0000000000..f32ac7023a
--- /dev/null
+++ b/site/docs/use-cases/2-Image-Generation/_sections/_run_model/_code_example_python.mdx
@@ -0,0 +1,13 @@
+import CodeBlock from '@theme/CodeBlock';
+
+
+{`import openvino_genai as ov_genai
+from PIL import Image
+
+pipe = ov_genai.Text2ImagePipeline(model_path, "${props.device || 'CPU'}")
+image_tensor = pipe.generate(prompt)
+
+image = Image.fromarray(image_tensor.data[0])
+image.save("image.bmp")
+`}
+
diff --git a/site/docs/use-cases/2-Image-Generation/_sections/_run_model/index.mdx b/site/docs/use-cases/2-Image-Generation/_sections/_run_model/index.mdx
new file mode 100644
index 0000000000..9b2bf5cc91
--- /dev/null
+++ b/site/docs/use-cases/2-Image-Generation/_sections/_run_model/index.mdx
@@ -0,0 +1,43 @@
+import CodeExampleCPP from './_code_example_cpp.mdx';
+import CodeExamplePython from './_code_example_python.mdx';
+
+## Run Model Using OpenVINOβ’ GenAI
+
+OpenVINO GenAI introduces `Text2ImagePipeline` for inference of text-to-image models such as Stable Diffusion 1.5, 2.1, XL, LCM, Flux, and more.
+See all supported [image generation models](/docs/supported-models/#image-generation-models).
+
+
+
+
+
+
+
+
+
+
+
+
+
+ :::info
+
+ Code below requires installation of C++ compatible package.
+ See [here](https://docs.openvino.ai/2025/get-started/install-openvino/install-openvino-genai.html#archive-installation) for additional setup details,
+ or [How to Build OpenVINOβ’ GenAI APP in C++](https://medium.com/openvino-toolkit/how-to-build-openvino-genai-app-in-c-32dcbe42fa67) blog for full instruction.
+
+ :::
+
+
+
+
+
+
+
+:::tip
+
+Use CPU or GPU as devices without any other code change.
+
+:::
diff --git a/site/docs/use-cases/2-Image-Generation/_sections/_usage_options/index.mdx b/site/docs/use-cases/2-Image-Generation/_sections/_usage_options/index.mdx
new file mode 100644
index 0000000000..9a5c86098a
--- /dev/null
+++ b/site/docs/use-cases/2-Image-Generation/_sections/_usage_options/index.mdx
@@ -0,0 +1,74 @@
+## Additional Usage Options
+
+:::tip
+Check out our [Python](https://github.com/openvinotoolkit/openvino.genai/tree/master/samples/python/) and [C++](https://github.com/openvinotoolkit/openvino.genai/tree/master/samples/cpp) samples.
+:::
+
+### Use Different Generation Parameters
+
+You can adjust several parameters to control the image generation process, including dimensions and the number of inference steps:
+
+
+
+ ```python
+ import openvino_genai as ov_genai
+ from PIL import Image
+
+ pipe = ov_genai.Text2ImagePipeline(model_path, "CPU")
+ image_tensor = pipe.generate(
+ prompt,
+ # highlight-start
+ width=512,
+ height=512,
+ num_images_per_prompt=1,
+ num_inference_steps=30,
+ guidance_scale=7.5
+ # highlight-end
+ )
+
+ image = Image.fromarray(image_tensor.data[0])
+ image.save("image.bmp")
+ ```
+
+
+ ```cpp
+ #include "openvino/genai/image_generation/text2image_pipeline.hpp"
+ #include "imwrite.hpp"
+
+ int main(int argc, char* argv[]) {
+ const std::string models_path = argv[1], prompt = argv[2];
+
+ ov::genai::Text2ImagePipeline pipe(models_path, "CPU");
+ ov::Tensor image = pipe.generate(
+ prompt,
+ // highlight-start
+ ov::genai::width(512),
+ ov::genai::height(512),
+ ov::genai::num_images_per_prompt(1),
+ ov::genai::num_inference_steps(30),
+ ov::genai::guidance_scale(7.5f)
+ // highlight-end
+ );
+
+ imwrite("image.bmp", image, true);
+ }
+ ```
+
+
+
+:::info Understanding Image Generation Parameters
+
+- `width`: The width of resulting image(s).
+- `height`: The height of resulting image(s).
+- `num_images_per_prompt`: Specifies how many image variations to generate in a single request for the same prompt.
+- `num_inference_steps`: Defines denoising iteration count. Higher values increase quality and generation time, lower values generate faster with less detail.
+- `guidance_scale`: Balances prompt adherence vs. creativity. Higher values follow prompt more strictly, lower values allow more creative freedom.
+- `rng_seed`: Controls randomness for reproducible results. Same seed produces identical images across runs.
+
+For the full list of generation parameters, refer to the [API reference](https://docs.openvino.ai/2025/api/genai_api/_autosummary/openvino_genai.Text2ImagePipeline.html#openvino_genai.Text2ImagePipeline.generate).
+
+:::
+
+### Working with LoRA Adapters
+
+TBD
diff --git a/site/docs/use-cases/2-Image-Generation/index.mdx b/site/docs/use-cases/2-Image-Generation/index.mdx
new file mode 100644
index 0000000000..7cd3500a23
--- /dev/null
+++ b/site/docs/use-cases/2-Image-Generation/index.mdx
@@ -0,0 +1,11 @@
+---
+sidebar_position: 1
+---
+import RunModelSection from './_sections/_run_model/index.mdx';
+import UsageOptionsSection from './_sections/_usage_options/index.mdx';
+
+# Image Generation Using Diffusers
+
+
- {`import argparse
-from PIL import Image
-import openvino_genai
-
-def main():
- parser = argparse.ArgumentParser()
- parser.add_argument('model_dir')
- parser.add_argument('prompt')
- args = parser.parse_args()
-
- device = 'CPU' # GPU, NPU can be used as well
- pipe = openvino_genai.Text2ImagePipeline(args.model_dir, device)
- image_tensor = pipe.generate(
- args.prompt,
- width=512,
- height=512,
- num_inference_steps=20
- )
-
- image = Image.fromarray(image_tensor.data[0])
- image.save("image.bmp")`}
-
-);
-
-const cppCodeBlock = (
-
- {`#include "openvino/genai/image_generation/text2image_pipeline.hpp"
-#include "imwrite.hpp"
-int main(int argc, char* argv[]) {
-
- const std::string models_path = argv[1], prompt = argv[2];
- const std::string device = "CPU"; // GPU, NPU can be used as well
-
- ov::genai::Text2ImagePipeline pipe(models_path, device);
- ov::Tensor image = pipe.generate(prompt,
- ov::genai::width(512),
- ov::genai::height(512),
- ov::genai::num_inference_steps(20));
-
- imwrite("image.bmp", image, true);
-}`}
-
-);
-
export const ImageGeneration = () => {
return (
@@ -75,8 +31,12 @@ export const ImageGeneration = () => {
- {pythonCodeBlock}
- {cppCodeBlock}
+
+
+
+