Pastaio

Create something new.

Choose your inputs and check the cost before you start.

1

Choose what to create

2

Start with your idea

Choose what you want to work with.

Explore ideas
Specific details help. You don’t need a perfect prompt.

Include your subject, lighting and composition.

Nano Banana Pro supports this creation method

Recent creations

Pick up where you left off.

Your creations will be kept here once you sign in

Set everything up now; you are asked to sign in when you press Create, and nothing here is lost.

Not sure what to make? See finished work and its prompts

Choose a model

Find the right tool for this task. Your compatible settings stay with you.

Text to image22 compatible modelsOnly models that support this mode are shown.
Nano Banana ProNewCurrent
Pastaio x Nano Banana

Google's Nano Banana Pro: still images at 1K, 2K or 4K from text or a picture, in ten frame shapes.

1k / 2k / 4k
≈25 credits
Nano Banana 2New
Pastaio x Nano Banana

Google's balanced high-volume image model.

1k / 2k / 4kLook things upFile type
≈12 credits
Nano Banana 2 Lite
Pastaio x Nano Banana

The Lite tier of Nano Banana 2: stills from text or a picture in ten frame shapes.

Model-defined size
≈6 credits
Seedream 4.5New
Pastaio x Seedream

ByteDance's image model — crowded, multi-character scenes without losing coherence.

Model-defined sizeDetail
≈10 credits
Flux Kontext Pro
Pastaio x Flux

Black Forest Labs' precise image editing and generation.

Model-defined size
≈6 credits
Flux Kontext Max
Pastaio x Flux

The larger Flux Kontext tier: pictures from text or from a picture.

Model-defined size
≈10 credits
Flux Kontext Dev
Pastaio x Flux

The Dev tier of Flux Kontext: stills from text or a picture in six frame shapes.

Model-defined size
≈4 credits
Flux 2 ProNew
Pastaio x Flux

Black Forest Labs' latest-generation image model.

1k / 2k
≈7 credits
GPT Image 1.5New
Pastaio x OpenAI

OpenAI's image model: stills from text or a picture in three frame shapes, including lettering in the picture.

Model-defined size
≈11 credits
Google Imagen 4 Ultra
Pastaio x Google

Google's Imagen 4 Ultra tier: still images from text, in five frame shapes.

Model-defined size
≈12 credits
Google Imagen 4 Fast
Pastaio x Google

Stills from text in five frame shapes.

Model-defined size
≈4 credits
Google Imagen 4
Pastaio x Google

The standard tier of Google Imagen 4: stills from text in five frame shapes.

Model-defined size
≈8 credits
Ideogram V3
Pastaio x Ideogram

Ideogram V3: stills from text or a picture, with lettering and graphic compositions; in Image editing it widens a picture to a new frame.

Model-defined sizeStyle
≈10 credits
Qwen Image 2.0 ProNew
Pastaio x Qwen

Alibaba's image model: stills from text or a picture in nine frame shapes, with multilingual lettering.

Model-defined size
≈18 credits
Qwen Image 2.0
Pastaio x Qwen

The standard tier of Qwen Image 2.0: the same nine frame shapes and multilingual lettering.

Model-defined size
≈8 credits
Grok Imagine
Pastaio x Grok

xAI's image model: stills from text or a picture in five frame shapes.

Model-defined size
≈10 credits
Grok Imagine Quality
Pastaio x Grok

Grok Imagine's other still mode: the same five frame shapes from text; it takes longer to render.

The same price as the standard mode; it takes longer.

Model-defined size
≈10 credits
Grok Imagine Image 2.0
Pastaio x Grok

xAI's newer image model; its pictures can be refined again by prompt.

Model-defined size
≈10 credits
Wan 2.7
Pastaio x Wan

Wan 2.7 stills: from text or a picture in nine frame shapes.

Model-defined sizeThink it through first
≈10 credits
Wan 2.7 Pro
Pastaio x Wan

The Pro tier of Wan 2.7's stills — 4K output, and it thinks the prompt through first.

Model-defined sizeThink it through first
≈20 credits
Wan 2.6
Pastaio x Wan

Stills from Wan 2.6 at a size you set, and prompt edits of a picture.

Model-defined sizeWidthHeight
≈8 credits
Kling O1
Pastaio x Kling

Kling's image model: stills and prompt edits at 1K or 2K, in eight frame shapes.

1k / 2k
≈8 credits
Only models that work with this way of starting are listed. The figures are rough guides; sign in to see your own price before anything starts.

Choose with confidence

Start with the job you need to do.

A model is the engine behind your result.

Different models support different inputs and controls. There is no single best model for every idea.

01

Choose the task first

Create from a prompt, animate an image, guide a scene with references, or transform an existing video. The list then shows only compatible models.

02

Check the controls you need

Look for your required duration, frame format, references and audio options. Settings are drawn from each model’s supported configuration.

03

Try a small, representative shot

Before committing to several creations, test the motion, visual style or voice that matters most. Review the quoted cost each time.

Keep creating

Choose a credit pack when you need more.

Available balance0 credits

You’ll be taken to a secure checkout. Credits are added to your balance once the payment is confirmed.

Creation details