What kind of photo should I upload?

Front-facing or 3/4 angle, face in focus, decent lighting, no sunglasses or mask covering most of your face. Chest-up framing works best. JPG or PNG, up to 20MB.

Will I still look like me?

Yes. Each style is built to keep your face shape, eyes, nose, mouth and skin tone intact and only swap the lighting, outfit and background. AI isn't 100% perfect — treat the result as a stylized portrait, not a literal you.

Are my photos saved or used to train AI?

No. We use uploads only for the current generation, then they're cleaned up. Never used to train any model. Don't upload someone else's photo without their permission.

Why do styles cost different credits?

The Top quality tier costs more, the Cheap tier costs less. Headshots and wedding shoots default to Top (detail matters); 3D figures and film portraits default to Recommended. The exact cost is shown before each generation.

Are the portrait styles free? What do paid credits unlock?

Free previews come with a small watermark — fine for testing. Once you've bought any credit pack you can download the clean, full-resolution original.

Ghibli Style Portrait Generator — Turn Your Photo Into a Hand-Painted Anime Still

Prompt

Redraw the uploaded photo as a still frame from a hand-painted Japanese animated feature film of the classic 1990s–2000s era.

CRITICAL: This must be the same person. Preserve the subject's facial identity exactly — face shape, eye shape, nose, mouth, hairline, facial hair, skin tone, ethnicity and apparent age must remain unmistakably the uploaded subject. Keep the original pose, gesture, clothing, props and composition; only the rendering changes. Do not add extra people.

Character rendering — this is the key: the person must look like a 2D animation cel, not a painted portrait. Clean, thin, slightly soft dark-brown outlines around the face, hair, hands and clothing; flat cel-shaded color fills with only one hard-edged shadow tone and one highlight; smooth, untextured skin; simple anime facial features (clean eye shapes with a small highlight, a minimal nose, simple mouth) while keeping the real face shape and proportions; hair as a few big clumped shapes with a handful of strands. Semi-realistic anime proportions true to the real body — no oversized eyes, no chibi or doll-like features. No brush strokes, watercolor bleed or paper texture on the person.

Background: a separate hand-painted background layer in gouache and watercolor — visible brush texture, simplified shapes, soft edges, rich but natural colors — so the crisp cel-drawn character stands out against the painterly set, like a frame from a hand-drawn animated film. Indoor scenes keep their walls, furniture and objects, simplified and hand-painted.

Light and color: warm late-afternoon sunlight raking in from one side, dappled window and leaf shadows across walls and floor (painted as clean shapes on the character), golden rim light on hair and shoulders, cool blue-grey shadows, slightly hazy atmosphere. Nostalgic, calm, cozy mood.

Output: 2:3 vertical frame, film-still composition, subject clearly readable.

Style-specific avoid: no photorealistic skin or photo texture, no realistic watercolor portrait of the person, no 3D render or CGI look, no glossy digital airbrushing, no heavy black outlines, no exaggerated big-eye moe face, no named film characters, no readable text or logos, no subtitles.

Avoid: explicit or suggestive content, revealing clothing, watermarks, text artifacts, extra limbs, deformed hands, face distortion. Keep every subject fully clothed and at their real age. The result must preserve the uploaded subject's exact facial identity — same person, just rendered in this style.

Prompt

Redraw the uploaded photo as a still frame from a hand-painted Japanese animated feature film of the classic 1990s–2000s era.

Output: 2:3 vertical frame, film-still composition, subject clearly readable.

Prompt

Redraw the uploaded photo as a still frame from a hand-painted Japanese animated feature film of the classic 1990s–2000s era.

Output: 2:3 vertical frame, film-still composition, subject clearly readable.

Model	Best For	Highlights	Credits/Image (Low–Top)
GPT Image 2	Everyday images, posters with text, quick drafts	Lowest cost at Low quality, sharp in-image text	10–110

Quality	Image Detail	Best For	GPT Image 2
Low	Good composition, lighter detail	Testing prompts, bulk drafts, social posts	10
Medium	Rich detail and textures	Marketing images, slides, blog art	40
High	Fine detail, clean text	Print, posters, product shots	110
Auto	The model picks a level	When you are unsure	40

Prompt

Redraw the uploaded photo as a still frame from a hand-painted Japanese animated feature film of the classic 1990s–2000s era.

Output: 2:3 vertical frame, film-still composition, subject clearly readable.

Ghibli Style Portrait

Here's roughly what your photo will look like

Three steps. No studio. No camera.

1. Pick a style

2. Upload your photo

3. Wait under a minute

Why this beats slapping a filter on it

Your face stays yours

Real range

Ready to post

Try free first

Your photo is safe

Browser only

FAQ

Pick a style. Upload one photo.

Looking for a different tool?

How to Use the GPT Image 2 Generator

How to Use the GPT Image 2 Generator