짧은 답변
제목과 크레딧을 위한 여백이 있는 포스터 사진.
짧은 답변
제목과 크레딧을 위한 여백이 있는 포스터 사진.
Local look lab
Upload a photo you own. The canvas, grade, grain, and caption stay in this browser.
Choose a photo to start. Nothing leaves this browser.
AI 스튜디오
이미지 모델은 사진, 로컬 에디터는 글자를 맡습니다. 깨진 글자를 피할 수 있습니다.
제목과 크레딧을 위한 여백이 있는 포스터 사진.
AI 스튜디오
Prompt lab
A strong photographic base makes the text layout easier.
Create an original vertical K-drama key-art photograph. A fictional young adult couple stands apart in a rain-soaked Seoul alley at night, lit by a single convenience-store sign and a cold blue street lamp. One holds a phone with a blank glowing screen; the other looks back over a shoulder. Use cinematic anamorphic photography, realistic rain, shallow depth of field, and a restrained teal-and-warm-red palette. Leave clean negative space at the top and bottom for a title. The people must be original fictional characters, not real actors or celebrities. No text, no logos, no watermarks, no extra people.
Avoid: real actors, celebrities, text, logos, watermark, extra people
A budget edit route is enough when the composition is simple.
Create an original vertical horror-mystery poster photograph. A fictional young woman stands in a dim apartment hallway, facing a phone that casts a cold light on her face. A wall clock behind her is out of focus; the hands are not readable. Use deep shadows, a narrow pool of light, realistic skin texture, and a muted green-and-red palette. Leave dark negative space at the top for a typographic title and at the bottom for credit blocks. The subject is an original fictional character, not a real actor. No text, no logos, no watermark, no extra people.
Avoid: real actor, text, logo, watermark, extra people
Character routes help keep two original faces distinct.
Create an original vertical romance-fantasy key art. A fictional couple stands on a rooftop above a sleeping city, with a large moon and thin clouds behind them. One wears a dark coat; the other wears a pale dress with a subtle star pattern. Use soft cinematic light, realistic fabric, and a deep blue-and-silver palette. Keep the faces small enough to feel like a poster but clear enough to read as two distinct people. Leave clean negative space at the top and bottom for typography. No text, no logos, no watermark, no third person.
Avoid: text, logo, watermark, third person, blurry faces
Model routing
| Provider | Best for | Typical route | Docs |
|---|---|---|---|
| falSpeed-first | Fast hosted queues and strong model coverage | FLUX, Kling, and other hosted media models | Docs |
| WaveSpeedBudget-first | Low-cost image edits and image-to-video drafts | Seedream edit, FLUX Kontext, Wan image-to-video | Docs |
| KieCharacter-first | Nano Banana image edits and Kling video routes | Nano Banana, Kling video | Docs |
| SeeAPIRouting layer | One API surface for several model families and fallback routes | Image and video inference endpoints | Docs |
| ReplicateExperiment-first | Broad catalogue for experiments and batch comparisons | Community and first-party image/video models | Docs |
AI 스튜디오
English
FAQ
아니요. 로컬 에디터는 브라우저 안에서 동작합니다. 공급자에게 프롬프트를 붙여 넣으면 해당 서비스 약관이 적용됩니다.
원본 사진에 대한 권리가 있고, 모델 라이선스가 상업적 사용을 허용하며, 필요한 곳에 AI 사용을 밝히는 경우에만 가능합니다. 타인의 얼굴은 동의 없이 쓰지 마세요.
확인한 출처