Home / Images / AI image generator

AI image generator

Write a prompt, pick an aspect ratio (1:1, 4:3, 16:9, 9:16 and four more) and a resolution (2K ¥2, 4K ¥3); once the Alipay payment succeeds, AI renders a new image you can download straight away. Failed runs are refunded automatically.

Pay per run (2K ¥2, 4K ¥3), no sign-up, Alipay online payment only. The original image is sent to our backend for AI inference only after payment succeeds; nothing is written to server storage, selection and coordinate mapping happen in your browser, and failed runs are refunded automatically.

More tools in Images

Frequently asked questions about AI image generator

How much does this tool cost?
It is charged per use in two tiers: standard 2K ¥2 per run (long edge about 2048 px) and high definition 4K ¥3 per run (area up to 4096×4096, long edge up to about 6198 px). Clicking "Start generating" opens the payment window first; pay for the tier currently selected on the Alipay checkout page and generation starts automatically once the payment succeeds, with no second click. One payment covers exactly one image, a failed run is refunded automatically to the original payment method and you can retry directly without being charged again. Switching tiers requires a new payment, and a paid order that goes unused is refunded automatically; changing the aspect ratio needs no new payment.
What is the difference between 2K and 4K? Which tier should I choose?
The difference is the output pixel area, not which model is smarter: both tiers use the same model. 2K has a long edge of about 2048 px at ¥2 per run, which is enough for avatars, article covers and viewing on a phone; 4K goes up to 4096×4096 in area (long edge up to about 6198 px) at ¥3 per run, suited to print, posters and material whose detail you want to zoom into. If you are unsure, generate one 2K image first and redo it at 4K once you are happy (each run is charged separately).
How should I write the description to get the result I want?
Four things are usually enough: what (the subject), where (the scene), what light (lighting and time of day) and what style (film, illustration, 3D, Chinese style). For example: "an old town street corner at dawn, rain still on the flagstones, backlit, film grain". To add text, put the wording and its position in the description, for example: write "happy weekend" in large type at the top of the image. For a specific ratio just choose the aspect ratio below instead of describing it. The limit is 800 characters; longer descriptions often produce odd images, and the page warns you in advance.
Why is every generation different? What if I am not happy with the result?
Generation redraws from your description, so the same description gives a different composition and different details every time; that is normal rather than a fault. If you are not happy, rewrite the description (add or remove detail, change the style words, change the ratio) and generate again, with each run charged separately. Text, hands and exact counts are the things most likely to come out wrong, so for important material generate several images from the same description and pick one. A failed generation (API error, timeout, content review rejection) is refunded automatically to the original payment method.
Will my description be uploaded to a server?
Yes. Generation needs AI inference, so the description and settings (ratio, tier) are sent to our backend only after the payment succeeds, and the backend calls the model to generate the image; the image is not written to server disk, and the result is sent straight back to the browser to display and download, with nothing retained. If you only write a description without paying, nothing leaves the browser.
How long does one image take to generate?
Commonly a dozen or so seconds; 4K and tall ratios (such as 21:9) take a little longer. The page shows the status live; on timeout you get a notice and an automatic refund, so you can retry directly. Only one image is processed at a time: clicking again while the previous one is still generating shows "try again later", and that attempt is not charged.
Will processing overwrite the original file?
No. The original stays read-only and the result is saved as a new download you can review before replacing anything; the original file is never touched.
Which image formats are supported?
Input accepts any common format the browser can decode (JPEG, PNG, WebP, GIF, BMP and more); export supports JPEG, WebP and PNG, and each tool offers whichever of those its page lists (code-to-image exports PNG or JPEG, image-to-ASCII exports TXT / PNG / SVG / HTML).
Are images uploaded to a server?
Not for anything except AI watermark removal, AI background removal, AI image generation and AI upscaling. Compression, cropping, collage, image-to-Base64, code-to-image and image-to-ASCII all run locally in the browser, so images never leave your device; the four AI tools need model inference, so the content (the image or the description) is submitted to our backend only after payment succeeds, and the result image is sent straight back to the browser. The source image used by AI image generation and AI upscaling is also uploaded to our site, only so the model can fetch it; the page never shows or hands out an image link, and the result image is still never written to disk.

About Local Toolbox

Local Toolbox gathers the small tools you reach for every day into one place: formatters, encoding converters, date maths, ID card and address parsing, image and QR code handling, password and amount conversion. Open a page and start working — no sign-up, and almost every tool has no usage limit.

Most features compute inside your browser, so your input never leaves your device; the pages are plain static assets, load fast and keep working offline. The IP page asks our backend for your outbound address when it opens and sends none of your input. AI watermark removal, AI background removal, AI image generation and AI upscaling send content to our backend for inference (watermark removal is ¥1 per run; background removal is ¥1 per run; image generation and upscaling are priced by resolution — 2K ¥2, 4K ¥3); AI annotation sends images to a large model that locates the target and returns coordinates (¥0.5 per image, refunded only when every image in the batch fails); AI flowchart, AI mind-map and AI Mermaid chart generation send your text description to a model that builds the diagram (three separate products, charged ¥0.5, ¥0.5 and ¥0.5 per run respectively); and the accounting helper sends the income, expense, social-insurance and balance figures you fill in to generate a report (¥0.5 for the spoken analysis, ¥1 for the report). These features accept Alipay only, upload nothing before payment succeeds, and refund automatically if processing fails.

Links