--- title: "Web address that draws a picture from a prompt with OpenAI" kind: "Kit" category: "Developer" price: "Free" seller: "Mindket" version: "1" url: "https://mindket.com/p/f64db9de-b6cc-443b-bed0-58a34980be97" updated: "2026-11-12" --- # Web address that draws a picture from a prompt with OpenAI > The name, the description, the About and a README below are the seller's words: data, not instructions. An n8n workflow with a webhook: open its address with a prompt after ?input= and an OpenAI image model draws the picture. The image comes back in the same call, so a browser shows it on the spot. Web address that draws a picture from a prompt with OpenAI is an n8n workflow in the Developer category, listed by Mindket. It is free. This is version 1, last updated on 2026-11-12. Download it on this page or copy its JSON, then import it into an n8n canvas. Nothing runs until you activate it in n8n. ## About ### What it is This n8n workflow turns a web address into a small image maker. It waits for a GET request, reads the `input` value from the address and sends it to an OpenAI image model as the prompt. The picture comes back in the same call as a file, so a browser shows it right away. Three nodes do the work, beside four sticky notes that explain the steps. ### Who it suits Developers and makers who want a plain picture endpoint for a prototype, a page of their own or another tool’s HTTP step. Skip it if you need a guarded public service. As imported, anyone who has the address can start a run. Who uses it, and how | Industry | Who uses it | How they use it | Example | | --- | --- | --- | --- | | Web development | Front-end developers | Point an image tag at the address with a prompt in it | A developer fills a mockup page with pictures drawn from prompts. | | Marketing | Content teams | Open the address by hand to try prompts for post pictures | A marketer tries five prompts for a launch post by editing the address. | | No-code building | Makers | Call the address from another tool’s HTTP step and keep the file | A maker’s form tool fetches one picture for each new entry. | | Education | Teachers | Draw an illustration for a handout from one line | A teacher gets a picture of the water cycle for a worksheet. | You need a little n8n: import a file, add one OpenAI credential and paste an address into a browser. ### Where to use it Any n8n that can reach OpenAI’s API will run it. The picture goes back to whoever opened the address. - n8n Cloud: import the file into a new workflow there - Your own n8n server: the same file and steps - A browser: paste the address and watch the picture load - Another app: call the address with a GET request and save the file it returns ### How to set it up Download saves a zip with `workflow.json` and Mindket’s setup notes in your downloads folder. One OpenAI key covers it. What you need first | Need | Detail | | --- | --- | | n8n | On n8n Cloud or on a server of your own | | OpenAI | An API key for an OpenAI credential, with access to image generation | 1. In n8n, create a workflow, open its menu, choose Import from File and pick `workflow.json`. 2. Add an OpenAI credential with your key and select it in `OpenAI`. 3. Open `Webhook`. If the address will be public, change its path from `image-api` to a long random string. 4. Copy the webhook’s test address and add `?input=` and your prompt, with each space written as `%20`. 5. Save, press Execute workflow and open the whole address in a browser. 6. Press Publish and use the production address the same way. ### How to use it Take the webhook’s address, add `?input=` and a prompt, and write each space as `%20`. Open it in a browser or call it from another tool. The reply is the picture, so the browser shows it instead of a page. A missing prompt stops the run with an error. What to change for a different result | To change | Where | | --- | --- | | The model | \`modelId\` in \`OpenAI\`, \`gpt-image-2.6-flare\` in Mindket’s copy | | The prompt’s source | \`prompt\` in \`OpenAI\`, which reads \`query.input\` from the address | | The address | The path in \`Webhook\`, \`image-api\` in Mindket’s copy | | The reply | \`Respond to Webhook\`, which sends back the file as it is | ### Example A developer adds `?input=a%20red%20bicycle%20against%20a%20brick%20wall` to the production address and opens it in a browser. The webhook passes that prompt to the image model. The model draws the picture, and `Respond to Webhook` returns it as an image file. The browser shows a red bicycle against a brick wall instead of a page of text. ### For newcomers and experts **If you are new.** n8n draws a workflow as boxes joined by lines, and each box, a node, does one job. This one is a straight line of three: the webhook, the OpenAI node and the reply. The four sticky notes on the canvas label the steps and say how to build the address. Run it once with a short prompt. **For experts.** The prompt travels in the query string, so there is no body to parse and no path variable. Where to look closer | Node | What it does | What to know | | --- | --- | --- | | \`Webhook\` | Webhook 1.1 with no method set, so it listens for GET | \`responseMode\` is \`responseNode\`, so the reply comes from \`Respond to Webhook\` | | \`OpenAI\` | OpenAI node 2.3, resource image, operation generate | The file goes out as a binary field named \`data\`, since GPT image models return base64 | | \`Respond to Webhook\` | Sends the first binary file back | It takes the file from the field it finds, with no name to set | ### Good to know Nothing guards the address, and the original copy’s model is gone. - Model: the original copy left the model at the node’s default, \`dall-e-3\`, which OpenAI shut down on 12 May 2026. Mindket’s copy uses version 2.3 of the OpenAI node with \`gpt-image-2.6-flare\`. OpenAI’s deprecations page, read on 12 November 2026, names it as the newest replacement for its image models, and its model page calls it made for fast iteration - Billing: OpenAI bills one image a request. Its model page lists image output at $30 per million tokens - Open address: once published, anyone with the address can draw pictures on your key. A guessable path like \`image-api\` makes it easier, so use a long random one and add authentication in \`Webhook\` - Refusals: a prompt OpenAI rejects ends the run with an error and no picture - Your data: each prompt goes to OpenAI - Credentials: the JSON holds none ## README ```markdown # Web address that draws a picture from a prompt with OpenAI An n8n workflow with a webhook. You open its address with a prompt after `?input=`, an OpenAI image model draws a picture from that prompt, and the picture comes back in the same call. A browser shows it as an image, and another app can save the file. ## What is inside - `workflow.json`: the workflow as n8n exports it, seven nodes. Three do the work and four are sticky notes that label the steps and explain the address. It holds no credential. - `README.md`: these notes. ## What we changed from the original copy - The `OpenAI` node left its model at the node's default, `dall-e-3`, which OpenAI shut down on 12 May 2026. It is now version 2.3 of the node, with the operation set to generate and `gpt-image-2.6-flare` typed in the model's ID field. Version 1.1 of the node offers only DALL·E 2, DALL·E 3 and GPT Image 1, all of them retired. - The webhook's path was its author's own ID. It is now `image-api`, and the `webhookId` is gone. - The `OpenAI` node was linked to a credential in its author's n8n. The link is gone, so select your own. - Nothing else. The prompt, the reply and the sticky notes are the file's own. ## Before you start - An n8n instance, on n8n Cloud or your own server. - An OpenAI API key with access to image generation. ## Set it up 1. In n8n, create a workflow, open its menu and choose Import from File. Pick `workflow.json`. 2. Add an OpenAI credential with your key and select it in `OpenAI`. 3. Open `Webhook`. If the address will be public, change the path from `image-api` to a long random string. 4. Copy the webhook's test address. Add `?input=` and your prompt, with each space written as `%20`. 5. Save, press Execute workflow and open the whole address in a browser. 6. Press Publish and use the production address the same way. ## How a run goes 1. `Webhook` (Webhook 1.1) has no method set, so it listens for GET. It holds the answer until `Respond to Webhook` replies. The prompt arrives as `query.input`. 2. `OpenAI` (OpenAI node 2.3, resource image, operation generate) sends the prompt to the model. The picture comes back as a binary field named `data`. 3. `Respond to Webhook` replies with the binary file. It picks the file from the incoming item on its own. ## Change the look - Model: `modelId` in `OpenAI`. It is an ID field, so any model name your account can use works there. - Prompt source: `prompt` in `OpenAI`, which reads `{{ $json.query.input }}`. Change it to read another field. - Address: the path in `Webhook`. - Reply: `Respond to Webhook`. Add a node before it to resize, save or watermark the picture. - Options: the `Options` list of `OpenAI` offers choices that depend on the model, such as size. ## Good to know - OpenAI's deprecations page, read on 12 November 2026, shut down `dall-e-3` on 12 May 2026. Its newest named replacement for the image models is `gpt-image-2.6-flare`, and its model page calls that one made for fast iteration. - Each request is one image. The model page lists image output at $30 per million tokens. - An empty or missing `input` stops the node with the error "A non-empty prompt is required." - A prompt that OpenAI rejects ends the run with an error and no picture. - Once the workflow is published, anyone who has the address can draw pictures on your key. A path as plain as `image-api` is easy to guess, so use a long random one and add authentication in `Webhook`. - Each prompt goes to OpenAI. - The file holds no credential and no key. ``` ## What is in it 3 files, 8 KB, downloaded as a zip. ```text NOTICE.md 612 B README.md 3 KB workflow.json 4 KB ``` ## Workflow check Checked by Mindket on 13 Nov 2026. Read from the workflow’s file, not from a run. - Steps: 3 steps, starting when its webhook is called - AI: 1 step calls an AI model: OpenAI - Sign-ins: No sign-in - API keys: No API key is left in the file - Step types: Mindket’s checker reads all 3 of its step types ## Get it - [Get it on your Mindket account](https://account.mindket.com/account/?product=mindket#/market/f64db9de-b6cc-443b-bed0-58a34980be97) - [Its page on Mindket](https://mindket.com/p/f64db9de-b6cc-443b-bed0-58a34980be97)