AI Image Generator

Turn a prompt or a reference photo into an image with Nano Banana, GPT Image and Seedream. Pick the model, press Generate, compare the results. Free to try, no sign-up.

No sign-up needed: guests can make 5 images a day, with Seedream 5.0 Flash only. Outputs are up to 1024 px with a small "pixmint.ai" watermark, for personal non-commercial use. The other 6 models need a free account (50 credits on first sign-in plus 20 every day).

Examples: one prompt, different models

Every image below was generated on Pixmint with the prompt printed under it. We ran each prompt once per model, so these are single samples, not a benchmark. Use them to get a feel for how the models interpret the same words.

Natural-light portrait of a woman by a cafe window, generated with Nano Banana 2
Portrait · Nano Banana 2Prompt: Natural-light portrait of a woman in her 30s by a cafe window, soft afternoon light, realistic skin texture, 50mm photo
White ceramic coffee mug product photo on a white background, generated with GPT Image 2.5
Product photo · GPT Image 2.5Prompt: E-commerce product photo of a white ceramic coffee mug on a pure white background, soft shadow, studio lighting
Yellow poster with the text SUMMER SALE 50% OFF, generated with GPT Image 2
Poster text · GPT Image 2Prompt: Minimal poster with the exact text "SUMMER SALE 50% OFF" in bold sans-serif on a sunny yellow background
Watercolor children's book illustration of a fox reading under a tree, generated with Seedream 4.0
Illustration · Seedream 4.0Prompt: Children's book illustration of a fox reading under a tree, watercolor, warm colors
Rainy Tokyo alley at night with lanterns and a cat under an umbrella, generated with Nano Banana Pro
Night scene · Nano Banana ProPrompt: Tokyo alley at night in the rain with lanterns and a cat under an umbrella, cinematic
Isometric 3D clay-style home office icon in pastel colors, generated with Seedream 5.0 Pro
3D icon · Seedream 5.0 ProPrompt: Isometric 3D icon of a cozy home office, soft pastel colors, clay render

How to generate an image in three steps

  1. Write a promptSay what the picture shows, where it happens and how it should look. A sentence or two is enough to start; you can write up to 3,000 characters.
  2. Choose a model, shape and countStart with Seedream 5.0 Flash, the free model, or open the model list for Nano Banana, GPT Image and the other Seedream versions. Pick an aspect ratio (1:1, 4:3, 3:4, 16:9 or 9:16) and generate one to four images at once.
  3. Review, refine and downloadDownload what you like, press Regenerate for another take, or use a result as a reference image for the next run. You can also send any result to the AI image editor to change a detail.

Which model should you pick?

There is no single best model. They trade price, speed and style, and the right choice depends on the job. This is a rough guide based on the example images on this site; open a model page for its full example set.

ModelMakerCreditsGood starting point for
Seedream 5.0 FlashByteDance5Quick drafts and free experiments.
Seedream 4.0ByteDance8Bright, decorative illustrations and posters.
Seedream 5.0 ProByteDance12Detailed scenes and bold typography.
GPT Image 2OpenAI15Clean graphic compositions and product shots.
GPT Image 2.5OpenAI15All-round work, including editing a photo.
Nano Banana 2Google18Photographic, documentary-style scenes.
Nano Banana ProGoogle36Rich detail when a prompt is complex.

Not sure? Generate the same prompt with the free model first. If the idea works, run it again with a stronger model for the final version. That costs a few cents instead of paying the highest price for every attempt.

The models page lists every model with prices, and the comparison pages put two models side by side, for example Nano Banana 2 vs GPT Image 2.5.

What people use an AI image generator for

Image generators are most useful when you need a visual quickly and do not have the budget or time for a photo shoot or an illustrator. Common jobs:

Social media and ads

Square, vertical and wide formats for posts, stories and banners. Generate four options, keep the best.

Product visuals

Clean studio-style product shots and lifestyle backdrops for listings, mockups and pitch decks.

Illustrations

Children's book scenes, blog headers, game concepts and icons in watercolor, flat or 3D styles.

Concept art and mood boards

Explore settings, lighting and atmosphere before committing to a direction.

Presentations and posters

Cover images and posters, including short headlines rendered as text inside the picture.

Variations of an existing image

Add up to four reference images to restyle a photo or build on a look. See image to image.

Prompt tips that actually help

  • Name the subject first, then the setting: "a red fox reading under an oak tree" works better than a list of adjectives.
  • Say how it is lit and shot: "soft afternoon light", "50mm photo", "studio lighting" give photographic results; "watercolor" or "clay render" gives an illustration look.
  • Put exact text in quotation marks if it must appear in the image, and keep it short. Check the spelling in the result every time.
  • Choose the aspect ratio before you generate. 9:16 for stories, 16:9 for banners, 1:1 for feeds and icons.
  • Change one thing per attempt. If you rewrite the whole prompt each time you cannot tell what made the difference.

More detail, with before-and-after prompt rewrites, is on the text to image page.

Pricing

Every model has a fixed price per image, shown on its button before you press Generate. Seedream 5.0 Flash is the cheapest at 5 credits ($0.05).

One credit is worth about $0.01 at the planned list price. The signup bonus of 50 credits and the 20 daily credits are enough to try several of the paid models, and the photo tools cost 15 credits per run.

ModelMakerCredits per imageList price per image
Seedream 5.0 Flash (free tier)ByteDance5$0.05
Seedream 4.0ByteDance8$0.08
Seedream 5.0 ProByteDance12$0.12
GPT Image 2OpenAI15$0.15
GPT Image 2.5OpenAI15$0.15
Nano Banana 2Google18$0.18
Nano Banana ProGoogle36$0.36

Planned paid plans (not available yet): 500 credits for $4.90, 1,100 credits for $9.90, 2,400 credits for $19.90, or a monthly plan of 1,200 credits for $9.90. Paid plans are planned but not available yet, so nothing can be bought today (see pricing).

Guests (no account) get 5 free Seedream 5.0 Flash images and 3 photo-tool uses per day. Signed-in users get 50 credits on their first sign-in plus 20 credits every day. Free outputs are up to 1024 px with a small watermark and are for personal non-commercial use.

Explore the models

AI image generator FAQ

Is this AI image generator really free?

You can try it without an account: guests get 5 free Seedream 5.0 Flash images per day. Signed-in users also get 50 credits on first sign-in and 20 daily credits to use on any model. Free images are up to 1024 px with a small watermark and for personal non-commercial use.

Which AI models does Pixmint use?

Seedream 5.0 Flash, Seedream 4.0, Seedream 5.0 Pro, GPT Image 2, GPT Image 2.5, Nano Banana 2, Nano Banana Pro. They come from Google, OpenAI and ByteDance. Pixmint is an independent service and is not affiliated with those companies.

Can I upload my own images?

Yes. Add up to four reference images (JPG, PNG or WebP, up to 20 MB each) and describe how the new image should use them. For changing one photo, the AI image editor is usually the better tool.

Can I use the images commercially?

Not today. Images made with free credits carry a watermark. Commercial use is planned to come with paid plans, which are not available yet, so today free outputs are for personal, non-commercial use only. See the terms and make sure your content does not infringe anyone else's rights.

What happens if a generation fails?

Failed images are refunded automatically and guest quota is given back. Blocked prompts are not charged.

How long are my images kept?

Guest images are deleted after 24 hours and free accounts' images after 7 days (30 days is planned for paid accounts, once they exist). Download anything you want to keep.

Start creating

Type a prompt at the top of this page and generate your first image.

Back to the generator