@inductiv/node-red-openai-api/src/images/help.html

Enhance your Node-RED projects with advanced AI capabilities.

<section>
    <details>
        <summary style="font-weight: bold;">🖼️ Images</summary>
        <a href="https://developers.openai.com/api/reference/typescript/resources/images/" target="_blank">Official Documentation
            <i class="fa fa-external-link fa-sm" aria-hidden="true"></i></a>
        <h4 style="font-weight: bolder;"> ⋙ Create Image</h4>
        <p>Creates an image given a prompt.</p>
        <dl class="message-properties">
            <h4>msg.payload Properties</h4>

            <dt>
                prompt
                <span class="property-type">string</span>
            </dt>
            <dd>A text description of the desired image(s).</dd>
            <dt class="optional">
                model
                <span class="property-type">string</span>
            </dt>
            <dd>
                The model to use for image generation. Supported model ids include
                <code>gpt-image-2</code>, <code>gpt-image-2-2026-04-21</code>,
                <code>gpt-image-1.5</code>, <code>chatgpt-image-latest</code>,
                <code>gpt-image-1</code>, <code>gpt-image-1-mini</code>,
                <code>dall-e-2</code>, and <code>dall-e-3</code>.
            </dd>
            <dt class="optional">
                n
                <span class="property-type">integer</span>
            </dt>
            <dd>The number of images to generate.</dd>
            <dt class="optional">
                quality
                <span class="property-type">string</span>
            </dt>
            <dd>The quality of the image that will be generated.</dd>
            <dt class="optional">
                response_format
                <span class="property-type">string</span>
            </dt>
            <dd>
                For <code>dall-e-2</code> and <code>dall-e-3</code>, the format in
                which generated images are returned. Must be one of
                <code>url</code> or <code>b64_json</code>.
            </dd>
            <dt class="optional">
                output_format
                <span class="property-type">string</span>
            </dt>
            <dd>
                For GPT image models, the output format. Must be one of
                <code>png</code>, <code>jpeg</code>, or <code>webp</code>.
            </dd>
            <dt class="optional">
                background
                <span class="property-type">string</span>
            </dt>
            <dd>
                For GPT image models, the requested background. Must be one of
                <code>transparent</code>, <code>opaque</code>, or <code>auto</code>.
                <code>gpt-image-2</code> and <code>gpt-image-2-2026-04-21</code> do
                not support transparent backgrounds; for those models use
                <code>opaque</code> or <code>auto</code>. Transparent backgrounds
                require <code>output_format</code> set to <code>png</code> or
                <code>webp</code>.
            </dd>
            <dt class="optional">
                size
                <span class="property-type">string</span>
            </dt>
            <dd>
                The size of the generated images. For
                <code>gpt-image-2</code> and <code>gpt-image-2-2026-04-21</code>,
                arbitrary <code>WIDTHxHEIGHT</code> strings are supported, for
                example <code>1536x864</code>. Width and height must both be
                divisible by 16, and the requested aspect ratio must be between 1:3
                and 3:1. Standard GPT image sizes include <code>1024x1024</code>,
                <code>1536x1024</code>, and <code>1024x1536</code>; <code>auto</code>
                is supported where the model allows it. <code>dall-e-2</code> keeps
                the fixed <code>256x256</code>, <code>512x512</code>, and
                <code>1024x1024</code> sizes. <code>dall-e-3</code> keeps
                <code>1024x1024</code>, <code>1792x1024</code>, and
                <code>1024x1792</code>.
            </dd>
            <dt class="optional">
                style
                <span class="property-type">string</span>
            </dt>
            <dd>
                The style of the generated images. Must be one of vivid or natural.
            </dd>
            <dt class="optional">
                stream
                <span class="property-type">boolean</span>
            </dt>
            <dd>
                If <code>true</code>, the node emits image stream events as they
                arrive instead of returning a final <code>response.data</code> array
                in one message.
            </dd>
            <dt class="optional">
                partial_images
                <span class="property-type">integer</span>
            </dt>
            <dd>
                When streaming GPT image responses, request between 0 and 3 partial
                images before the completed image event.
            </dd>
            <dt class="optional">
                user
                <span class="property-type">string</span>
            </dt>
            <dd>
                A unique identifier representing your end-user, which can help OpenAI to
                monitor and detect abuse.
            </dd>
        </dl>

        <h4 style="font-weight: bolder;"> ⋙ Create Image Edit</h4>
        <p>
            Creates an edited or extended image given an original image and a prompt.
        </p>
        <dl class="message-properties">
            <h4>msg.payload Properties</h4>

            <dt>
                image
                <span class="property-type">string | string[]</span>
            </dt>
            <dd>
                The image to edit. Provide one local file path, or an array of local
                file paths for GPT image edit requests. GPT image models can accept up
                to 16 images; <code>dall-e-2</code> still expects a single image.
            </dd>
            <dt>
                prompt
                <span class="property-type">string</span>
            </dt>
            <dd>A text description of the desired image(s).</dd>
            <dt class="optional">
                mask
                <span class="property-type">string</span>
            </dt>
            <dd>
                An additional image whose fully transparent areas (e.g. where alpha is
                zero) indicate where image should be edited. If you send multiple
                input images, the mask applies to the first image.
            </dd>
            <dt class="optional">
                model
                <span class="property-type">string</span>
            </dt>
            <dd>
                The model to use for image generation. Use <code>dall-e-2</code> or a
                GPT image model such as <code>gpt-image-2</code>,
                <code>gpt-image-2-2026-04-21</code>, <code>gpt-image-1.5</code>, or
                <code>chatgpt-image-latest</code>.
            </dd>
            <dt class="optional">
                n
                <span class="property-type">integer</span>
            </dt>
            <dd>The number of images to generate. Must be between 1 and 10.</dd>
            <dt class="optional">
                size
                <span class="property-type">string</span>
            </dt>
            <dd>
                The size of the generated images. <code>gpt-image-2</code> and
                <code>gpt-image-2-2026-04-21</code> support arbitrary
                <code>WIDTHxHEIGHT</code> strings such as <code>1536x864</code> when
                width and height are divisible by 16 and the aspect ratio stays between
                1:3 and 3:1.
            </dd>
            <dt class="optional">
                response_format
                <span class="property-type">string</span>
            </dt>
            <dd>
                For <code>dall-e-2</code>, the format in which the generated images are
                returned. Must be one of <code>url</code> or <code>b64_json</code>.
            </dd>
            <dt class="optional">
                output_format
                <span class="property-type">string</span>
            </dt>
            <dd>
                For GPT image models, the output format. Must be one of
                <code>png</code>, <code>jpeg</code>, or <code>webp</code>.
            </dd>
            <dt class="optional">
                background
                <span class="property-type">string</span>
            </dt>
            <dd>
                For GPT image models, the requested background. Must be one of
                <code>transparent</code>, <code>opaque</code>, or <code>auto</code>.
                <code>gpt-image-2</code> and <code>gpt-image-2-2026-04-21</code> do
                not support transparent backgrounds.
            </dd>
            <dt class="optional">
                stream
                <span class="property-type">boolean</span>
            </dt>
            <dd>
                If <code>true</code>, the node emits image edit stream events as they
                arrive instead of returning a final <code>response.data</code> array.
            </dd>
            <dt class="optional">
                partial_images
                <span class="property-type">integer</span>
            </dt>
            <dd>
                When streaming GPT image edits, request between 0 and 3 partial images
                before the completed image event.
            </dd>
            <dt class="optional">
                user
                <span class="property-type">string</span>
            </dt>
            <dd>
                A unique identifier representing your end-user, which can help OpenAI to
                monitor and detect abuse.
            </dd>
        </dl>

        <h4 style="font-weight: bolder;"> ⋙ Create Image Variation</h4>
        <p>Creates a variation of a given image.</p>
        <dl class="message-properties">
            <h4>msg.payload Properties</h4>

            <dt>
                image
                <span class="property-type">string</span>
            </dt>
            <dd>The image to use as the basis for the variation(s).</dd>
            <dt>
                model
                <span class="property-type"></span>
            </dt>
            <dd>The model to use for image generation.</dd>
            <dt>
                n
                <span class="property-type">integer</span>
            </dt>
            <dd>The number of images to generate.</dd>
            <dt>
                response_format
                <span class="property-type">string</span>
            </dt>
            <dd>The format in which the generated images are returned.</dd>
            <dt>
                size
                <span class="property-type">string</span>
            </dt>
            <dd>The size of the generated images.</dd>
            <dt>
                user
                <span class="property-type">string</span>
            </dt>
            <dd>
                A unique identifier representing your end-user, which can help OpenAI to
                monitor and detect abuse.
            </dd>
        </dl>

    </details>
</section>