Recommended Free Tools
Windows Errors? Fix Them Before They Spread
Free scan · no reinstall
Crashes, No Sound, or Screen Glitches?
Do these 3 things before closing this tab:
1Scan for outdated or missing drivers - takes under a minute2Repair Windows errors before they cause bigger problems3Fix the driver behind crashes, sound loss and screen glitchesFree scan · under a minute
PC Slower Than It Used to Be?
Free scan · Windows 10 & 11
ThatPainter is reader-supported. When you buy through links on our site, we may earn an affiliate commission. Learn More
You do not need a special ChatGPT command or a permanent “Studio Ghibli” preset to make this kind of image. Describe the qualities you want—such as hand-painted backgrounds, delicate linework, expressive faces, warm natural light, atmospheric depth, and a quiet sense of wonder—and refine the result with targeted requests.
| # | Preview | Product | Price | |
|---|---|---|---|---|
| 1 |
|
Hayao Miyazaki | $31.49 | Buy on Amazon |
| 2 |
|
The Art of Spirited Away | $28.00 | Buy on Amazon |
| 3 |
|
The Art of Howl's Moving Castle | $16.59 | Buy on Amazon |
| 4 |
|
The Art of My Neighbor Totoro: A Film by Hayao Miyazaki | $15.53 | Buy on Amazon |
| 5 |
|
The Art of Princess Mononoke | $18.59 | Buy on Amazon |
ChatGPT Images 2.0 is available on all ChatGPT tiers on the web, iOS, and Android. The separate Images with thinking capability is currently available on Plus, Pro, and Business; Enterprise and Edu access is listed as coming later. Use image generation to develop an illustration from a written idea, transform a reference photo, or revise an image. OpenAI’s Images in ChatGPT help page describes the current workflow.
Free tools Windows power users keep installed
One-click scans. No signup required.
What you need before you start
Decide what the image is for and collect any references you are allowed to use. A useful brief answers six questions:
#1 Best Overall
- Purpose: Is it a portrait, landscape study, poster, book illustration, or social-media image?
- Subject: Who or what is the focal point?
- Action: What is happening in the scene?
- Setting: Where and when does it take place?
- Visual treatment: What materials, colors, lighting, and atmosphere should the image have?
- Format: Should it be square, portrait, landscape, or another aspect ratio?
OpenAI Academy’s image-generation guidance says one to three clear sentences are usually sufficient. A specific description will generally produce a more controllable result than simply writing “make this Ghibli style.”
How to create the first image
On the web
- Open ChatGPT and select More in the sidebar.
- Choose Images, then select Create Image.
- Alternatively, ask ChatGPT to create an image directly in any conversation.
- Enter a detailed prompt and wait for generation to finish. OpenAI says complex requests may take a few minutes, and you can continue using ChatGPT while an image is generating.
On a phone or tablet
Open the composer’s tool controls, which may appear under the sliders-style Skills icon, then choose Create an image. The precise placement can vary between app versions. There is no required /imagine command; OpenAI describes image creation using ordinary natural-language prompts. See the ChatGPT release notes for the mobile Skills menu.
A strong prompt for a Ghibli-inspired mood
For a distinctive result, describe the visual ingredients instead of depending on a studio or brand name. Try this:
Create a warm, hand-painted Japanese animated-film-inspired illustration of a child riding a bicycle through a quiet countryside village at golden hour. Use expressive character features, soft watercolor backgrounds, gently textured brushwork, lush greenery, whimsical small details, atmospheric depth, and a nostalgic, wonder-filled mood. Avoid logos, copyrighted characters, recognizable movie scenes, and text.
This wording establishes the subject, action, setting, materials, lighting, mood, and restrictions. It asks for a general visual direction rather than an exact imitation of a particular studio, artist, film, or character. OpenAI’s image guidance recommends describing generic or ownable designs and avoiding brand references where possible.
Prompt examples for painters
Landscape
Create a hand-painted animated-film-inspired landscape of a small cottage beside a mossy stream in rural Japan. Show layered hills fading into atmospheric haze, tall grasses moving in a light breeze, soft watercolor textures, fine ink-like contours, dappled morning light, and a calm, gently magical mood. Landscape composition, 16:9 aspect ratio, no people, logos, or text.
Rank #2
- New
- Mint Condition
- Dispatch same day for order received before 12 noon
- Guaranteed packaging
- No quibbles returns
Portrait
Create a whimsical hand-painted animated-film-inspired portrait of a young woman standing beside a sunlit window. Preserve a natural, recognizable expression, use delicate linework, expressive eyes, soft watercolor textures, warm rim light, muted botanical colors, and a richly painted background with subtle atmospheric depth. Vertical portrait composition, no text or logos.
Interior scene
Create a cozy hand-painted animated-film-inspired kitchen in an old countryside house. Show wooden shelves, copper pots, jars of herbs, afternoon sunlight, floating dust motes, worn painted surfaces, soft watercolor washes, delicate contours, and many small but believable domestic details. Warm, peaceful, storybook atmosphere, 4:3 aspect ratio, no lettering.
Turning a photo into an illustrated portrait
Upload the image with the composer’s + or attachment button and choose Add photos & files. You can also drag an image into the text area or paste one from the clipboard. See OpenAI’s image inputs FAQ for supported upload methods.
Use a prompt that identifies what must stay recognizable:
Transform this photo into a whimsical hand-painted animated-film-inspired portrait. Keep the person’s facial identity, pose, clothing, and background composition recognizable, while using soft watercolor textures, expressive eyes, warm natural light, delicate linework, and a richly painted countryside atmosphere. Do not add text or logos.
Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Only upload photographs you have the right to use. OpenAI recommends getting permission to use a real person’s likeness, particularly when relevant to your intended use.
Rank #3
- New
- Mint Condition
- Dispatch same day for order received before 12 noon
- Guaranteed packaging
- No quibbles returns
How to revise without losing the character
Do not rewrite the entire prompt for every correction. Broad revisions can change the face, hairstyle, clothing, pose, or camera angle. Make one targeted request at a time and state what should remain fixed.
Useful follow-up instructions include:
Keep the composition and character design unchanged. Make the lighting warmer, simplify the background, and add more atmospheric haze.
Keep everything else exactly the same. Change only the sky to a soft sunset gradient and reduce the saturation of the foreground.
Preserve the person’s face, clothing, pose, and camera angle. Change only the rendering style to softer watercolor textures with delicate hand-drawn linework.
Repeating important constraints is worthwhile. OpenAI recommends targeted changes and explicit instructions such as “change only X” and “keep everything else exactly the same.” Image generation is not a pixel-perfect editing process, so these instructions cannot guarantee that every other detail will stay untouched.
Editing a particular area
- Open the generated or uploaded image.
- Select Select and highlight the area you want to alter.
- Describe the change in the prompt.
- Use Undo, Redo, or Cancel if the result is not useful, and choose Save when it is acceptable.
You can also skip the selection tool and describe the edit in the conversation panel. The highlighted region is only an approximate target: OpenAI warns that edits can extend beyond the selected area. For example, when changing a jacket, say that the face, hands, hairstyle, pose, background, and lighting must remain unchanged.
Rank #4
- New
- Mint Condition
- Dispatch same day for order received before 12 noon
- Guaranteed packaging
- No quibbles returns
Using multiple reference images
A small set of references is easier to control than a large collection. Give each image a job and refer to it by number:
Image 1 is the photo whose composition and subject I want to preserve. Image 2 is a color and texture reference. Apply the visual qualities of Image 2 to Image 1 while keeping the subject, layout, and objects from Image 1 unchanged. Keep the warm ochre palette, soft paper texture, and pale green shadows from Image 2.
Use spatial descriptions such as left, right, foreground, and background when combining references. This is more precise than saying “use the second image’s look.”
Getting the composition and output size right
Specify the intended format in the prompt, or change it later using the editor’s Aspect ratio control.
| Use | Prompt direction |
|---|---|
| Phone wallpaper | “Vertical portrait composition, 9:16 aspect ratio; leave quiet space near the top.” |
| Landscape painting study | “Wide landscape composition, 16:9 aspect ratio; keep the horizon in the upper third.” |
| Square post | “Centered square composition, 1:1 aspect ratio; keep the subject clear of the edges.” |
| Printable art | “Portrait format with generous margins and no text.” |
For a poster or label, keep the wording short and put the exact text in quotation marks. Specify its placement, color, size, and font treatment. OpenAI recommends spelling uncommon words letter by letter when accuracy matters. Generated lettering can still contain errors, especially in dense layouts, so proofread it and consider finishing the typography in a dedicated graphics application.
Do these 3 things before closing this tab:
1Clear out junk files and repair common Windows errors2Fix the driver behind crashes, sound loss and screen glitches3Repair Windows errors before they cause bigger problemsCommon problems and practical fixes
| Problem | Likely cause | What to try |
|---|---|---|
| The result looks like generic anime. | The prompt names a style but gives no visual construction. | Add materials, lighting, line quality, background treatment, atmosphere, palette, and a concrete setting. |
| The face or clothing changes after revision. | The follow-up request is too broad. | Tell ChatGPT to preserve the face, hairstyle, clothing, pose, camera angle, and layout, then change one element only. |
| A local edit spills into nearby areas. | The selection tool is not perfectly precise. | Select a rough region and explicitly list everything that must remain unchanged. |
| The image contains unwanted lettering. | The prompt leaves text unspecified. | Say “no text, logos, labels, or watermarks,” unless lettering is part of the design. |
| The generation is slow. | Complex scenes and detailed instructions can take longer. | Wait a few minutes; OpenAI says you can continue using ChatGPT while the image is generating. |
| The request is refused or the result is unexpected. | Safety checks apply to prompts, references, and outputs. | Use references you have permission to use and describe an original scene rather than a recognizable copyrighted character or movie frame. ChatGPT Images applies safety checks to prompts, uploaded images, and generated outputs. |
Saving, finding, and deleting the image
Images created with ChatGPT are automatically saved in the Images area. Open an image and use Copy, Save, or Share; on the web, Save downloads it.
Best Value
Do not confuse Images with Library. Generated images continue to appear in Images, while uploaded and created files may also be stored in Library. To remove an image from My images, delete the conversation that created it. On the web, hover over the conversation title, choose •••, and select Delete. On iOS or Android, press and hold the conversation title and choose Delete. See OpenAI’s Library help page for more on file storage.
Publishing or selling the result
OpenAI says that, subject to its Terms and Content Policy, users own ChatGPT output and may reprint, sell, or merchandise it. This does not guarantee copyright protection in every jurisdiction or establish that an image is legally non-infringing. You remain responsible for your prompt, reference images, likenesses, trademarks, and other rights involved. See OpenAI’s explanation of ChatGPT and its sharing and publication policy.
For public sharing, manually review the image and make clear that it is AI-generated. OpenAI says images generated with ChatGPT include C2PA metadata and SynthID watermarks. Metadata can be stripped, and watermark detection is not guaranteed after every transformation; see OpenAI’s C2PA and SynthID help page.
The Tool Desk
Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →FAQ
Do I need DALL·E 3 or a special command?
No. ChatGPT Images 2.0 is the current image-generation experience. Open Images and choose Create Image, or ask for an image in an ordinary conversation. You do not need /imagine syntax. The separate DALL·E GPT remains available, but it is not required for this workflow.
Is there a permanent Studio Ghibli preset in ChatGPT?
OpenAI’s current documentation does not list a Studio Ghibli-specific button or preset. Describe the qualities you want—hand-painted watercolor textures, delicate linework, expressive faces, warm natural light, atmospheric backgrounds, and a whimsical mood—instead of relying on a studio name.
Which ChatGPT plans include Images with thinking?
The dossier lists the separate Images with thinking capability as available on Plus, Pro, and Business, with Enterprise and Edu access coming later. General ChatGPT Images 2.0 availability is listed across all ChatGPT tiers.
Can I change the aspect ratio after generating an image?
Yes. Open the image editor and use the Aspect ratio control, or specify the desired ratio in a new prompt. You can request formats such as 1:1, 9:16, 4:3, or 16:9, along with composition instructions that keep the subject in a useful position.
Recommended Free Tools
Can I sell an image made with ChatGPT?
OpenAI says users may reprint, sell, and merchandise output subject to its Terms and Content Policy. You are still responsible for permissions, uploaded references, likeness rights, trademarks, and applicable copyright law. Review the image yourself and make clear that it is AI-generated when sharing it publicly.
The Bottom Line
The best results come from treating ChatGPT as an image-making assistant rather than a one-click style filter. Give it a specific subject and composition, describe the paint and light, request an original animated-film-inspired treatment, then make small revisions with explicit preservation instructions. That workflow produces more coherent art and avoids depending on an unofficial preset or an unreliable shortcut.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
