AlreadyRan

你的创作工作室

AI 图片生成器

描述它，打磨它，让它成为你的。

正在加载模型…
登录后生成

Made with AlreadyRan

Real results from the generator, with the prompts that made them. Click one to watch it full size.

Overview

AI image generator

Describe the picture, choose a model, and generate. Add reference images when you want to edit a photo or keep a subject consistent.

Text to image, and image to image

Write what you want to see and the model renders it. Models that accept reference images can also work from your own pictures: restyle them, change one element, or use them as a guide for composition and character.

Each model exposes its own controls — aspect ratio, resolution, quality, number of images — and only the ones it genuinely supports.

Know the cost before you generate

You see the estimated credits before you click, together with the most that can be reserved. When the run finishes you are charged for what it actually used, and whatever was reserved on top of that comes straight back.

Keep what works

Every image is saved to your history with the exact prompt, model and settings that produced it. Reuse loads them back into the form so a good result is one click away from a variation.

Use cases

What people make with it

  • Product shots

    Put a product from your photo into a clean studio scene or a lifestyle setting.

  • Posters and covers

    Some models set short text inside the picture — handy for posters, thumbnails and covers.

  • Edit your own photo

    Change the background, the style or one element, and keep the rest of the picture.

  • Consistent characters

    Give several references of the same person or character to keep them recognisable.

  • Concept art

    Try looks fast: several variations of one idea in a single run where the model allows it.

  • Layers and cut-outs

    Some models return transparent backgrounds or separate layers, ready for your own design.

How it works

How to generate an image

  1. Step 1

    Choose a model from the picker at the top of the form.

  2. Step 2

    Write a prompt; add reference images if the model supports them.

  3. Step 3

    Set size and quality, check the credit estimate, and click Generate.

Models

Image models you can use

Every model shows only the settings it really supports. Pick one in the box above; switching keeps your prompt.

ModelGood atWhat it takes
Seedream 4.0BytePlusUnified 4K generation and editing1K/2K/4KImage input≤15 outputs
Seedream 4.5BytePlusConsistent multi-image editing with typography2K/4KImage input≤15 outputs
Seedream 5.0 FlashBytePlusFast low-cost image generation and editing1K/1.5K/2KImage inputTransparentPNG/JPEG
Seedream 5.0 Flash LayerizeBytePlusLayer splitting into base and PNG1K/1.5K/2KImage inputJPEG/PNG
Seedream 5.0 LiteBytePlusDeep-thinking generation with online search2K/3K/4KImage input≤15 outputsWeb searchPNG/JPEG
Seedream 5.0 ProBytePlusHigh-density infographic image generation1K/1.5K/2KImage inputTransparentPNG/JPEG
Seedream 5.0 Pro LayerizeBytePlusHigh-fidelity editing with layer splitting1K/1.5K/2KImage inputJPEG/PNG
Nano Banana ProGoogleStudio-quality 4K visuals with text rendering1K/2K/4KImage inputWeb search
Nano Banana 2GoogleHigh-efficiency generation and editing0.5K/1K/2K/4KImage inputWeb search
Nano Banana 2 LiteGoogleUltra-low latency 1K image generation1KImage input
Nano Banana 2.1Google4K output with search grounding1K/2K/4KImage inputWeb search
GPT Image 1.5OpenAIImproved prompt adherence for image generationImage input
GPT Image 2OpenAIPrompt-faithful generation with legible text1K/2K/4KImage input≤10 outputsTransparentPNG/JPEG/WEBP
GPT Image 2.5 FlareOpenAIFast everyday high-quality image generation1K/2K/4KImage input≤10 outputsTransparentPNG/JPEG/WEBP
GPT Image 2.5 SunburstOpenAIPrecise editing for image generation1K/2K/4KImage input≤10 outputsTransparentPNG/JPEG/WEBP
Grok Imagine Image 2.0xAIEditing with multiple reference images1K/2KImage input≤10 outputs
Midjourney V8.2MidjourneyAesthetic-first HD image generationImage input
Qwen Image 3.0AlibabaFaster bulk generation of text-rich images1K/2KImage input≤6 outputs
Qwen Image 3.0 ProAlibabaComplex newspaper and storyboard layouts1K/2KImage input≤6 outputs
Qwen Image EditAlibabaBilingual text editing and semantic editsImage input≤6 outputs2 tasks
Wan 2.5 ImageAlibabaEnhanced aesthetics and realistic textureImage input≤4 outputs2 tasks

FAQ

Frequently asked questions

Didn't find your answer? Email us at support@alreadyran.com.

Which model should I choose?

Hover a model in the picker to see what it is good at and which inputs it accepts. When in doubt, start with the default and compare a second model on the same prompt — your prompt is kept when you switch.

Can I edit my own photo?

Yes, with any model that accepts reference images: upload the photo and describe the change.

What happens if a generation fails?

The credits reserved for it come back, unless the model's provider charged for the failed attempt: then you are charged only for what the model actually used, and History says so. A failure on our side is never charged.

Ready to make something?

Write a prompt at the top of this page and see the credits before you generate.