Skip to content

AI Image Generator (Text to Image & Image Editing)

Try for free

Make images from text prompts and edit images with Gemini image models: restyle, change backgrounds, add or remove objects, combine photos. Batch prompts, aspect ratios, 1K to 4K, PNG/JPEG/WebP. One row per image with a download link. No API key needed.

🎨 AI Image Generator (Text to Image & Image Editing)

Turn text prompts into images and edit your own pictures with Google's Gemini image models: restyle a photo, swap a background, add or remove objects, or combine several images into one. Run one prompt or hundreds in a batch and get every image as a PNG, JPEG or WebP file with a download link, its size and the prompt it came from. No API key, no GPU and no login: you pay $3.40 per 1,000 images plus the AI model usage of each image.

✨ Why use AI Image Generator?

  • 🖼️ Text to image and editing in one place: describe a new image, or add up to several input images and describe the change.
  • 📦 Batches: many prompts and several variations per prompt in one run, made in parallel.
  • 📐 The shape you need: 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, 5:4, 4:5 and 21:9, from 1K up to 4K with the Gemini 3 models.
  • 🔤 Readable text in images: posters, labels and logos with spelled-out words (best with Gemini 3.1 Flash Image and Gemini 3 Pro Image).
  • 🔗 Ready-to-use files: every row has a download link that opens without an account, plus width, height, format and file size.
  • 💸 Pay per image: requests that return no image are not charged as an image.

🎯 Use cases

  • 🛍️ E-commerce: product shots on new backgrounds, lifestyle scenes and seasonal variants from one photo.
  • 📣 Marketing and social media: ad creatives, 9:16 stories and 4:5 feed posts, many variations to test.
  • 📰 Blogs and newsletters: header images and illustrations for every article in one batch.
  • 🎮 Concept art and mood boards: quick drafts of scenes, characters and styles.
  • 🏠 Real estate and interiors: restyle a room photo, try furniture and colour ideas.
  • 🤖 Automation: call it from your own code or a workflow to make images on demand.

🚀 How to use

  1. Write what you want under Prompt (and more prompts under More prompts for a batch).
  2. To edit pictures, paste their links (or pasted image data) under Images to edit and describe the change in the prompt.
  3. Pick the Model, Aspect ratio, Resolution and File format, and set Images per prompt for variations.
  4. Click Start. Each image appears as a row with a preview and a download link.
  5. Download the files from the links, or export the table as JSON, CSV or Excel.

📥 Input

FieldWhat it does
promptDescribe the image, or the change to make to the input images.
promptsMore prompts, one per line; each gets its own images.
imagesPerPromptVariations per prompt (default 1).
inputImageUrlsImages to edit: links (JPEG, PNG, WebP, GIF up to 30 MB) or data:image/...;base64,.... Empty = text to image.
modelgoogle/gemini-3.1-flash-lite-image (default, fastest and lowest cost), google/gemini-2.5-flash-image, google/gemini-3.1-flash-image, google/gemini-3-pro-image.
aspectRatioauto, 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, 5:4, 4:5, 21:9.
imageSize1K (default), 2K, 4K (2K and 4K with the Gemini 3 models).
negativePromptThings the image should not contain.
outputFormatpng (default), jpeg or webp.
maxConcurrencyImages made at the same time, 1 to 8 (default 4).

numImages, imageUrl and imageUrls are accepted too.

{
  "prompt": "A minimalist poster with the words \"SCRAPE.LY\" in bold letters over mountains",
  "prompts": ["A watercolor hummingbird on a pink flower"],
  "aspectRatio": "16:9",
  "imageSize": "2K",
  "model": "google/gemini-3.1-flash-image",
  "outputFormat": "jpeg"
}

Editing example:

{
  "prompt": "Show the cat from the first image and the cat from the second image sitting together on a green sofa, natural light",
  "inputImageUrls": [
    "https://upload.wikimedia.org/wikipedia/commons/thumb/4/4d/Cat_November_2010-1a.jpg/500px-Cat_November_2010-1a.jpg",
    "https://upload.wikimedia.org/wikipedia/commons/thumb/a/a3/June_odd-eyed-cat.jpg/500px-June_odd-eyed-cat.jpg"
  ],
  "model": "google/gemini-2.5-flash-image",
  "imagesPerPrompt": 2,
  "aspectRatio": "4:5"
}

📤 Output

One row per image:

FieldDescription
imageUrlDownload link of the image (opens without an account for 7 days).
promptThe prompt the image was made from.
modetext-to-image or image-to-image.
modelThe model that made the image.
width, heightSize in pixels.
aspectRatioThe aspect ratio asked for (auto when none).
format, contentType, fileSizeBytesFile format, MIME type and size.
promptNumber, imageNumberWhich prompt of the run and which variation of it.
inputImageUrlsThe images that were edited (empty for text to image).
negativePromptWhat the image was asked to avoid.
modelTextText the model wrote next to the image, if any.
fileKey, createdAtKey of the file in the run's storage, and when it was made (UTC).

Example row from a test run (link shortened):

{
  "imageUrl": "https://api.scrape.ly/v2/key-value-stores/82cc0e3f-92c4-4de0-b95f-b43246c0b51a/records/image-0001-001.jpg?signature=…&attachment=1",
  "prompt": "A minimalist poster with the words \"SCRAPE.LY\" in bold letters over mountains",
  "mode": "text-to-image",
  "model": "google/gemini-3.1-flash-image",
  "width": 2752,
  "height": 1536,
  "aspectRatio": "16:9",
  "format": "jpeg",
  "contentType": "image/jpeg",
  "fileSizeBytes": 534948,
  "promptNumber": 1,
  "imageNumber": 1,
  "inputImageUrls": [],
  "negativePrompt": null,
  "modelText": null,
  "fileKey": "image-0001-001.jpg",
  "createdAt": "2026-11-12T21:28:28Z"
}

💰 Pricing

  • Per image: $3.40 per 1,000 images on the Free plan, $3.23 on Starter, $2.97 on Scale and $2.55 on Business.
  • AI model usage is billed on top for each image, at the selected model's rate. With the default model (Gemini 3.1 Flash Lite Image, 1K) our test runs used about $0.04 per image on a paid plan (the Free plan has a higher AI usage rate, about $0.065 per image). The larger models and 2K/4K images cost more per image.
  • Compute for the run is billed too; it is small (well under $0.001 per image in our tests).

Example: 100 images with the default model on a paid plan ≈ $0.32 for the images + about $4 of AI model usage.

❓ FAQ

Do I need an API key or my own AI account? No. The models run through Scrape.ly; their usage is billed to your Scrape.ly account.

Can I edit my own photos? Yes. Paste image links or pasted image data under Images to edit and describe the change. Several images can be combined into one scene.

How many images can I make? As many as you like: add more prompts or raise Images per prompt. Large batches are made in parallel (up to 8 at a time).

What if an image is missing? Sometimes a model answers with text instead of an image, for example when it declines a prompt. The actor asks again once; if there is still no image, that request is logged as failed and is not charged as an image.

Can I cap what a run costs? Yes. Set a maximum cost for the run: the actor measures what each image costs and does not start an image that would go over it.

Can I run it on a schedule or from my code? Yes. Schedule it in the console, or start it through the Scrape.ly API, SDKs or integrations and read the rows from the dataset.

Who can use the images? You are responsible for your prompts, the images you upload and how you use the results. The models refuse some content.

⚠️ Limits

  • One image per request; variations are separate requests.
  • 2K and 4K need Gemini 3.1 Flash Image or Gemini 3 Pro Image; the other models make images of about 1 megapixel.
  • Input images larger than 2048 pixels are scaled down before they are sent; each must be under 30 MB.
  • Download links open without an account for 7 days. The files stay in the run's storage for your plan's data retention period.
  • Results are not reproducible: the same prompt gives a different image each time.
  • Free accounts have a small daily AI usage allowance; paid plans have none.

Related actors

//From the blog\\

Latest articles

All articles
G2
4.7/5
Excellent
Trustpilot

The internet is full of data.
Start owning yours.

Scrape.ly gives you the infrastructure, the tooling, and the team to extract, process, and operationalize web data at any scale. Setup takes minutes.

Get Started FreeChat with Support
No credit card. No setup calls.