Image to image
Edit a photo you already have
Upload an image and describe one change, such as a new season, background or color, while the subject stays recognizable. Most models accept a source image for edits.
AI Image Generator
Describe the image you want or upload a photo to change, then generate it with 20 AI models from OpenAI, Google, ByteDance, Black Forest Labs and more. New accounts get 10 free credits.
- Text to image
- Image to image
- Up to 16 reference images
- Up to 4K output
- One workspace
- Switch models on the same page.
- Upfront pricing
- See credits before you generate.
- Automatic refunds
- Failed runs return credits.
Prompt examples
AI image prompt examples you can copy
Each example shows a finished image with a prompt that describes it. Copy one into the workbench, change the subject and keep the structure: subject, setting, light and style.
Glass vase in hard sunlight
Example prompt: Product photo of a hand-blown amber glass vase holding orange ranunculus on a white ledge, hard sunlight, bright caustic patterns rippling across the wall, a crisp shadow and visible bubbles in the glass.
Bilingual shop poster
Example prompt: Retro print poster: a giraffe in a striped cape sits in a barber chair while a hedgehog on a wooden ladder trims its mane with scissors, mint tiled wall and a barber pole. A red banner at the bottom carries the title "长颈鹿理发店" and the small line "THE GIRAFFE BARBER".
Low-angle street portrait
Example prompt: Low-angle photo of a silver-haired woman in a lime green velour tracksuit and dark sunglasses power-walking toward the camera with pink dumbbells, a white wall and clear blue sky behind her, midday sun, natural skin texture, 24mm lens.
Clay character mid-action
Example prompt: Claymation-style frog chef in a white hat tossing a pancake out of a frying pan, pastel pink kitchen with a round window and a small yellow stove, soft studio light, fingerprints visible in the clay.
Two-color linocut print
Example prompt: Two-color linocut print of a heron with its wings spread wide, snatching a fish from red splashing water, bold carved lines, black and vermilion ink on cream paper, visible paper grain.
Which AI image models can you use here?
Switch models in the workbench above, or open a model page for examples, prompts and prices. The newest models come first.
New
GoogleNano Banana 2.1
Consistent characters, product placement, exact short text
View model
New
Black Forest LabsFLUX 3 Image
Composed scenes from a prompt or references
View model
New
IdeogramIdeogram 4.5
Refine image details with source-guided edits and readable text.
View model
OpenAIGPT Image 2.5 Flare
Faster everyday generation, natural lighting and rich textures
View model
OpenAIGPT Image 2.5 Sunburst
Precision-first detail, premium visual fidelity
View model
xAIGrok Imagine 2
Bold visual hooks, scene text, plain-language edits
View model
MicrosoftMAI Image 2.5 Pro
Object consistency, brand scenes, visual reasoning
View model
AlibabaQwen Image 3
Long prompts, multilingual typography, text-heavy images
View model
Also available
Seedream 5 Pro
Infographics, dense layouts, multilingual design
Nano Banana 2 Lite
Google's fastest Nano Banana for 1K images
MAI Image 2.5
Precise local edits, text updates, identity preservation
GPT Image 2
Readable text, complex briefs, precise edits
Nano Banana 2
Sentence-based edits, multi-image composites, in-image translation
Z-Image Base
Open styles, prompt control, visual studies
FLUX.2 Klein
Fast 4B and 9B drafts and low-cost photo edits
Seedream 4.5
Multi-image composition and reference fidelity
Z-Image Turbo
Fast variations at the lowest cost
FLUX.2 Pro
Photoreal products, materials, camera and lighting
Nano Banana Pro
Labeled diagrams, five-person shots, logos on products
Beyond text to image
What else can you do besides text to image?
Edit a photo you already have, combine several references into one scene, set exact words on signs and posters, or keep the same people across a series.
Up to 16 references
Combine reference photos into one scene
Add a person, a pet and a place as separate references, then say how they come together. The workbench shows how many references the selected model accepts.
Readable text
Put readable text on signs and posters
Write the exact words in quotes and say where they go. Text-focused models such as GPT Image, Ideogram, Qwen Image and Seedream keep short headlines, labels and multilingual signs legible.
Consistent characters
Keep the same people across a series
Upload a photo of your characters and ask for new scenes. Reference-based models such as Nano Banana keep faces, outfits and colors close, so a set reads as one story.
How it works
How to generate an AI image from text
Four steps from a written idea to a downloaded file.
Describe the image
Write the subject, setting, light and style in full sentences. Put the details that must not change first.
Add a photo if you have one
Upload an image to edit it, keep a face or product, or match a color palette. Skip this step for pure text to image.
Pick a model and size
Choose the model, aspect ratio and resolution. The Generate button shows the credit cost before you start.
Generate and download
Check faces, hands and any text at full size, then download as PNG, JPEG, or WebP.
Use cases
What can you create with an AI image generator?
Common jobs people bring to the generator, each with a short example brief.
Product photos
Put a product reference on new surfaces and in new light for listings, ads and launch pages.
Example: A ceramic mug on a linen tablecloth in morning window light, with space for a price tag.
Posters and social graphics
Make event posters, menus and post graphics with short readable headlines, in square, story or banner ratios.
Example: A 9:16 story poster for a Saturday night market titled "Night Bites".
Characters and game art
Design a character once, then show them in new poses and scenes for a comic, a campaign or a tabletop game.
Example: A dwarf blacksmith character sheet with front, side and back views.
Interior and home ideas
Upload a photo of a room and try new paint, furniture or materials before you buy anything.
Example: Paint this living room sage green and swap the sofa for a pink velvet one.
Explore more AI image tools
Remove the background, upscale, vectorize or denoise any result, all on the same credit balance.
Background Remover
Transparent PNG or WebP with clean hair and product edges, ready for a listing or a new composition.
1 credit per megapixelOpen tool
Image Upscaler
2× / 4× / 8× with recovered texture instead of smeared pixels, so a 1K draft can become a print.
0.25 credits per megapixelOpen tool
Image Vectorizer
Logos, icons and flat illustrations become editable SVG that scales to any size.
2 credits per imageOpen tool
Image Denoiser
Remove grain and compression artifacts from low-light photos and scans while keeping real detail.
16 credits per imageOpen tool
FAQ
AI image generator FAQ
Free credits, model choice, photos as input, text in images and more.
Is this AI image generator free?+
You can start free. New accounts get 10 credits, valid for 365 days, and images start at 1 cr on the cheapest model. You need an account to generate.
Which AI image model should I use?+
Pick by the job: GPT Image 2.5 Flare for fast everyday images, Nano Banana 2.1 for edits and consistent characters, Ideogram 4.5 or Qwen Image 3 for text-heavy designs, FLUX.2 Pro for photoreal products and Z-Image Turbo for cheap drafts. All 20 models run in the same workbench, and the model list above links to each model's examples.
Can I generate an image from a photo?+
Yes. Upload one or more images, up to 16 depending on the model, and describe what to keep and what to change. Use it to edit a photo, restyle a room or place a product in a new scene.
Can AI put readable text in an image?+
Yes, for short text. Put the exact words in quotes and say where they go, for example a red banner at the bottom that reads "OPEN DAILY". GPT Image, Ideogram, Qwen Image and Seedream handle headlines, labels and multilingual signs best. Check the spelling at full size before you publish.
How do I make AI images look realistic?+
Describe it like a photo: camera angle, lens, time of day, light source and natural skin or material texture. Skip stacks of words like "hyperrealistic 8K". Photoreal models such as FLUX.2 Pro, Z-Image Turbo and GPT Image 2.5 Flare are good places to start.
Can I keep the same character across several images?+
Yes. Upload a clear image of the character as a reference and describe each new scene. Nano Banana 2.1 and other reference-based models keep faces, outfits and colors close; use the same model for the whole set.
How do I write a good prompt?+
Write full sentences in this order: subject, setting, composition, light, style and where the image will be used. Put details that must not change first, then change one thing per attempt. The prompt examples above follow this structure.
What sizes and file formats can I download?+
Up to 4K, in 7 aspect ratios from square to wide banner, as PNG, JPEG, or WebP. The options shown depend on the selected model, and downloads have no visible watermark.
Are my images private?+
Images made on the free plan appear in the public gallery. Paid plans can keep results private.
Can I use the generated images commercially?+
Generally yes, and we do not claim ownership of the images you create here. Whether a specific image is safe to publish still depends on the selected model provider's terms, your plan, applicable law, and the rights attached to your prompt and reference images. Clear permission for identifiable people, brands, copyrighted artwork, and third-party product imagery, and take your own legal advice for high-stakes commercial use.
Start free
Generate your first AI image
New accounts get 10 free credits. Write a prompt or copy an example to begin.