Image
AI Style Transfer
AI style transfer applies the look of one image to another: its palette, lighting, texture, and treatment. Upload the image you want to transform, choose a style or describe one, and generate a new version that helps keep the main subject recognizable. On the Astorie canvas you can also supply the style as a reference image and keep it connected, so the same look can be applied to further images without setting it up again.
Free to start. No card required.
Try style transfer right now
Pick a source image and a style, hit Create, and watch the transfer render in place, using the same generation pipeline the canvas uses, on a real project you can keep editing.
Before and after examples
Before
After
Before
AfterWhat this feature solves
Prompt words are a weak way to describe a look. "Cinematic", "painterly", and "editorial" mean different things to different models, and the same words return a different result from one run to the next. A reference image gives the model something concrete to read: the palette, the lighting direction, the texture, and the way edges and shadows are handled.
The second problem is keeping the subject. A style pass that changes the product label, the face, or the packaging is not usable however good the look is. Style transfer works better when the source image and the style reference are supplied as two separate inputs, so the model can be pointed at what to keep and what to change. Tools with a single reference slot force both roles into one image and tend to weaken each.
The third is repeat work. One good styled image is easy. Twelve product shots across three aspect ratios, with a follow-up drop next month, is where most tools break down, because the look has to be described again in every new session and drifts a little each time.
How the workflow works in Astorie
- 1
1. Upload the source image
Add the photograph, product shot, or render you want to restyle. Use the highest-resolution copy you have; detail that is not in the source cannot be recovered later.
- 2
2. Add the style reference
Upload a second image carrying the look you want: palette, lighting, texture, and treatment. One coherent reference works better than a collage of several different styles.
- 3
3. Pick the model
Flux Kontext and Nano Banana 2 read a subject image and a style image as separate inputs and are the usual starting point. GPT Image 2 is useful for a cleanup pass afterwards.
- 4
4. Generate and review
Write a short prompt for anything the references do not already show, generate, and compare the result against the source. Adjust the reference before running the rest of the set.
- 5
5. Download or keep editing
Export the result, or continue on the canvas: upscale it, remove the background, or apply the same style reference to the next source image.
What you need to know
- Accepted files
- JPEG, PNG and WebP. Photos in HEIC or HEIF are converted to JPEG in your browser before they upload. GIF, animated WebP and SVG are not accepted.
- Image size
- Both sides must be at least 300 px. PNG and WebP files must be 30 MB or smaller. Larger JPEG files, and very large images of any accepted type, are resized in your browser before they upload.
- Uploading before you sign in
- You can attach an image before you have an account. It stays in this browser, up to 25 MB, and is uploaded only when you start a generation.
- What you get back
- One PNG image at 1K resolution, in a 1:1 frame.
- Batch generation
- This tool runs one generation at a time, and each run returns a single result.
- What it costs
- Uses up to 20 credits per image. Plan discounts and included allowances may lower this.
- Free allowance
- A new account starts on the free plan with 40 credits a month. Some accounts are asked to claim that allowance once, on the credits page, before it can be spent. See pricing for what the paid plans include.
- Your files and results
- Uploads and generated results belong to the project they run in. This demo creates that project in your own account rather than in a team workspace. How long they are kept, and how to have them deleted, is set out in the privacy policy.
- Commercial use
- What you may do with a result is governed by the terms, and the model you pick may carry its own restrictions. Read both before you publish or sell a result.
Why Astorie is different
On the Astorie canvas the source image and the style reference sit in separate labeled nodes and both are wired into the same image model. Flux Kontext and Nano Banana 2 read them as distinct inputs, which helps the model distinguish the style reference from the subject it should preserve. Pasting both into one slot gives the model no way to tell the two roles apart.
Once a look works it can be reused. Connect the same style reference to further source images and to square, vertical, and horizontal outputs, then save the finished setup so the next campaign only changes the source image. That is the part a single-generation tool cannot carry between sessions.
Example workflow
A fragrance brand has twelve bottles photographed on a neutral surface and one painterly editorial image that defines the campaign look. The style reference goes on the canvas once. Each bottle photograph is added as its own source image and wired into a Flux Kontext node alongside that reference, with a short prompt covering framing and background. The team generates one bottle first, checks that the label is still legible and the palette matches the reference, then runs the remaining eleven plus the square, vertical, and horizontal variants. Selected results go through an upscale pass before delivery, and the setup is saved so the next seasonal drop only needs new bottle photographs.
Common use cases
Apply one editorial look across a product line
Style each SKU photograph with the same treatment so the catalog reads as one shoot rather than a set of unrelated images.
Give a content series a recognizable look
Use a single style reference across a recurring post or episode format so each drop is identifiable before the caption is read.
Turn a mood board into finished assets
Take the reference image that defined the brief and apply it to the photographs the campaign actually needs.
Convert photography into illustration
Transform existing photos into a painterly, graphic, or illustrated treatment without re-shooting them.
Produce the same look at several sizes
Generate square, vertical, and horizontal versions of a styled image for in-feed, stories, and hero placements.
Recommended model stack
flux-kontext
imageReads a source image and a style reference as separate inputs, which helps preserve the subject while the treatment changes.
nano-banana-2
imageHandles several references at once and holds fine product and packaging detail well.
gpt-image-2
imageUseful for a cleanup pass on a styled result: edges, small artifacts, and text legibility.
midjourney
imageWide creative range when the look is described rather than supplied as a reference.
flux
imageHigh-fidelity output for hero placements that will be printed or upscaled.
Tips and common mistakes
Tips
- Use a source image with one clear main subject. A cluttered source makes it harder for the model to tell the subject from the background.
- Pick a style reference with a single coherent look. A reference mixing three unrelated styles usually produces an unstable result.
- Describe palette, lighting, texture, and composition in the prompt rather than leaning on one word such as "cinematic".
- Test the style on one image before running a whole set. Fixing the reference is cheaper than regenerating twenty outputs.
- Only upload images you have the rights to use, as a source and as a style reference.
Common mistakes
- Combining the subject and the style into a single reference image. The model has no way to tell which role each part plays.
- Relying on prompt words alone for the look. A reference image is a more stable input than a style adjective.
- Naming a living artist or a known studio. Keep style references generic and use imagery you have the rights to.
- Using a blurry or heavily compressed source. Style transfer tends to amplify existing artifacts rather than hide them.
- Scaling the run across a whole set before reviewing a single test output.
Frequently asked questions
Can I use any image as a style reference?
Any image you have the rights to use will work as an input, but a reference with one clear, coherent look gives a more predictable result than a mixed mood board. Avoid naming a living artist or a known studio in the prompt, and build references from your own imagery or from material you are licensed to use.
Will the subject stay recognizable?
Models like Flux Kontext and Nano Banana 2 are built to apply a style while preserving the subject, and supplying the source and the style as separate inputs helps them tell the two apart. Generation is still probabilistic, so review each result against the source before using it, particularly where a label, a face, or packaging text has to stay accurate.
Which image formats can I upload?
JPG, PNG, and WebP upload directly. HEIC and HEIF photos are converted to JPEG in the browser before upload. Animated and vector files are not accepted: GIF, animated WebP, and SVG are rejected.
Is there a size or resolution limit for uploads?
Both sides of an upload have to be at least 300 pixels. The 30MB ceiling is a hard limit for PNG and WebP only: files above it are rejected rather than recompressed, so reduce those before uploading. Larger JPEGs are re-encoded in the browser instead, and very large images of any accepted type are downscaled before upload.
What format is the download?
The generated image downloads in the format the model produced, usually PNG or JPEG. Astorie streams the original file rather than re-encoding it, so what you save matches what was generated.
Are the images I upload private?
Uploads and generated results are stored against your account. How they are processed, how long they are kept, and how to delete them are set out in the Astorie privacy policy, which is the authoritative source; this page does not restate its terms.
Can I use the results commercially?
Astorie does not claim ownership of the images you generate. Whether a specific result can be used commercially also depends on the rights in the source and reference images you uploaded and on the terms of the model that produced it. Check the Astorie terms of service before publishing or selling a result.
Is there a free allowance?
Astorie has a free plan that does not require a credit card and comes with a monthly credit allowance. Some accounts are asked to claim that allowance once, on the credits page, before it can be spent. Generation cost varies by model and output size, so the current allowance and per-model pricing are listed on the Astorie pricing page.
Can I style several images in one run?
Most image models on Astorie can return more than one variation per generation, and a group of nodes on the canvas can be run as a single batch, which is how one style reference is applied across a set of source images. The number of variations available per run depends on the model.
How is this different from character reference and character consistency?
This page covers applying the visual style of one image to another. If the goal is keeping a specific person recognizable from a portrait, see ai-character-reference. If the goal is the same character across a set of images, see ai-character-consistency. Style transfer changes how an image looks; those two are about who is in it.
Related how-to guides
Related models and tools
Tool
AI Image Upscaling
Upscale AI images to 4K, 8K, or print DPI on Astorie — sharpen keyframes before video generation and polish hero stills without leaving the canvas.
Tool
AI Background Removal
Remove backgrounds and cut out subjects for ecommerce, compositing, and video on Astorie — then recompose them into new scenes on the same canvas.
Provider
Google's Veo video, Imagen image, and Nano Banana model workflows on Astorie.
Provider
OpenAI
OpenAI's GPT Image and Sora video model workflows available on Astorie.
Provider
ByteDance
ByteDance's Seedance video and Seedream image model families on Astorie.
Related features
AI Character Reference — Reference-Image Workflows on Astorie
Use reference images to guide AI model outputs on Astorie's canvas.
AI Character Consistency Across Images and Video
Keep a subject consistent across image and video generations on Astorie using reference workflows.
AI Product Photography — Studio-Quality Product Images on Astorie
Generate studio-quality product photos for e-commerce on Astorie's canvas.
AI Photo Restoration — Restore Old Photos on Astorie
Restore old, damaged, or low-quality photos with AI on Astorie's canvas.
AI Headshot Generator — Professional Headshots in Minutes
Generate professional headshots for LinkedIn, resumes, and team pages on Astorie's canvas.
AI Mockup Generator — Product, Device, and Brand Mockups
Generate product, device, and brand mockups for marketing on Astorie's canvas.
AI Logo Generator — Brand Marks and Wordmarks on Astorie
Generate logo concepts, brand marks, and wordmarks on Astorie's canvas.
AI Emoji Generator — Custom Emoji on Astorie
Generate custom emoji and stickers for Slack, Discord, and brand on Astorie.
AI Sticker Generator — Telegram, WhatsApp, Discord Packs
Generate sticker packs for Telegram, WhatsApp, Discord, and iMessage on Astorie.
AI Comic Strip Generator — Multi-Panel Comics on Astorie
Generate multi-panel comic strips with consistent characters on Astorie's canvas.
AI Presentation Slides — Pitch Decks and Slide Visuals
Generate slide visuals, pitch deck imagery, and presentation graphics on Astorie.
AI Icon Generator — App and UI Icons on Astorie
Generate app icons, UI icons, and brand icon sets on Astorie's canvas.
AI Architecture Rendering — Building and Space Visualization
Generate architectural renderings, exterior visualizations, and concept art on Astorie.
AI Interior Design — Room and Space Visualization on Astorie
Visualize interior designs, room concepts, and decor schemes on Astorie's canvas.
AI Game Asset Generator — Sprites, Concept Art, Backgrounds
Generate game-ready assets, sprites, concept art, and backgrounds on Astorie.
Related docs
Related reading
Comparisons
Build it on the canvas
Open Astorie and wire this workflow up in minutes. Free to start — no card required.