How do you get AI to generate a different featured image for every WordPress post?

In short: Give the model a different kind of scene for each post and show it what the latest images in the same category looked like, with an instruction to do something else. Wiozen Content Bot does this for every article: it picks one of 8 scene types not used in the category's last 5 posts, gives the model the recent image descriptions, crops the image to 16:9 and puts the focus keyword in the alt text. Image generation is switched on in the Writing tab and billed separately from the text.

Why do AI images all look alike?

Give an image model only the post title and it draws the most likely scene every time. On a finance site that's a calculator, papers and glasses on a desk; on a food site, a plate on a wooden table. Ten posts later the home page looks like ten copies of one photo. The model doesn't remember earlier images; every request is its first.

Variety takes two things: changing the type of scene from outside, and telling the model what was drawn recently. The Content Bot does both inside the writing request.

How does the bot change the scene for each post?

The bot has 8 scene types: a close-up of the topic's main object, a moment with a person whose face isn't shown, a wide shot of the place where the topic happens, an overhead flat-lay on a plain surface, an outdoor scene in natural surroundings, one symbolic object on a plain colored backdrop, a flat illustration with a limited palette, and a simple 3D render in a soft clay style.

For each category the bot keeps the last 8 scenes and image descriptions it used. A new post can't get any scene used in that category's last 5 posts; one of the rest is picked at random. The model then has to state the style (photo, illustration or 3D) to match that scene.

The bot also hands the model the descriptions of the last 5 images in the same category and asks for a clearly different main object, place, colors and camera angle. The image history is per category: images in two different categories aren't compared with each other.

How does the image stay on topic?

The model that writes the article also writes the image description, in English, in the same request. The prompt tells it to show the concrete object, place or moment from the topic: a car or a dealership for a car loan post, the moment of paying by card for a credit card post, the dish itself for a recipe. A desk with a calculator, papers, a notebook, a pen and glasses is ruled out by name.

Money, license plates, signs and places match the article's language. A Turkish article should show Turkish lira and streets in Turkey, not another country's money or plates.

The image shouldn't contain recognizable real people, re-creations of real events or brand logos. The bot also adds a note to every image request asking for no text, letters, logos or watermark. So the Content Bot doesn't make cover images with the title or a logo on them; if you want that, edit the image in a design tool afterwards.

How are alt text, file name and size handled?

Along with the image description, the model writes short alt text. If the focus keyword isn't in it, the bot puts the keyword in front ("mortgage: a couple receiving the keys"). The same text becomes the image title in the media library. The file name is built from the post's slug and ID.

The bot asks for a 1536×1024 image and crops it to 16:9 from the center when saving it. Even if the model ignores the size and returns a square, the featured image ends up landscape. If the server has no WordPress image editor, the image is kept as it came.

How do I turn image generation on, and what does it cost?

1. Open the Content Bot in your WordPress dashboard and go to the Writing tab.
2. Under Image and publishing, tick Generate a featured image for each article (it's on by default in a new install).
3. Pick the image model from the list below it and save.

The default model is gemini-3.1-flash-image; as of October 30, 2026 the list also has gpt-image-2.5. Images are billed separately from articles and come off your Wiozen balance. At that day's prices, one image at the size the bot asks for costs about $0.047 with Gemini and about $0.116 with GPT Image 2.5. If Gemini can't answer at the moment, another model makes the image and you pay that model's price. Prices and that fallback are covered in the image generation cost guide.

What happens to the post if the image fails?

If the category publishes live and the image can't be generated, the post stays in drafts and the image goes into a queue. The bot retries after about 30 minutes, 2 hours, 6 hours and 15 hours; if the image is added within 5 attempts in total, it publishes the post. Failures caused by an invalid key or an empty balance don't use up attempts. If all 5 fail, the post stays a draft and the Log tab says so.

Waiting posts are listed on the Overview tab, each with a Retry now button, and you can retry all of them at once.

If the category saves drafts, a post whose image failed isn't queued. Those, along with drafts written before image generation was switched on, are counted on Overview as bot drafts without an image. The Generate missing images button adds an image to them from a prompt built from each draft's title and summary, handles up to 20 drafts per click and doesn't publish them.

Frequently asked questions

The Content Bot is free. Download the plugin, switch on image generation in the Writing tab, and look at the first run's images in the media library.

Create your account

More guides: How can you use an AI API from Turkey without a credit card? · How do you connect an OpenAI-compatible API to a tool you already use? · How do you calculate the cost of an AI API? · What should I do if my AI API key has leaked? · How do you generate images with an AI API, and what does one image cost? · Why does my AI API return a 429 error, and how do I fix it? · Why do long AI API requests time out: streaming or non-streaming? · Anthropic Messages API or OpenAI Chat Completions: which one should I use? · What is a context window, and what is a 1 million token context good for? · Which AI model should I choose: deciding by price, context and quality · Why does a reasoning model return an empty answer, and why are thinking tokens charged? · Why is my API balance running out faster than expected? Tracking spend request by request · How do you convert a web page to Markdown to feed it to an AI model? · How do you get AI to write and publish articles in WordPress automatically? · How do you find blog topics from RSS feeds and Google Trends? · Does AI-written content show up on Google and in AI answers? · How do you get internal links into AI-written WordPress articles? · How do you write a content brief when an AI writes your blog posts? · What should you watch for when an AI writes recipes for your food blog? · How do you fix errors when connecting a WordPress plugin to your AI API account? · Why does my AI article bot stop halfway in WordPress? "Run is stuck" and LiteSpeed · Why are my scheduled WordPress posts late? WP-Cron and a real server cron · How do I get an AI article bot to write blog posts in English or another language? · What does "Content Bot cannot generate articles" mean? Reading the bot's log