Editor’s Note:
I’ve been trying this kind of prompt style lately, and it does work pretty well.
But honestly, it’s not so much about the “structured” format itself. What really matters is having enough information in the prompt.
With ZIT and the Qwen text encoder, prompt adherence is very strong and the model doesn’t freestyle that much, so the amount of detail you put into the prompt makes a big difference in the final result.
When I get some time, I’ll put together a short post summarizing some prompt tips based on my recent experience with ZIT.
------
Qwen is a much smarter text encoder than the previous ones and it understand structure better than others. So I tried a structured method of prompting and it works wonders. IMO it's much easier to tweak than lengthy paragraphs and essays for prompts.
Photo1
Style:
\- 1970's Black and White Noir movie
\- Ultra high quality
\- Ultra high resolution
\- Close Up
\- Dim Lighting
\- Heavy shadows
\- Dramatic Lighting
\- Angled shot
\- Perspective shot
\- Depth of Field
Characters:
\- John, a 50 yr old man wearing a fedora and brown trench coat. He has a light stubble and weary face
Setting:
\- Art Deco style Streets of New York at night
Scene:
\- John is lighting standing and lighting a cigarette. The light from his lighter is illumninating his face.
\- At the bottom it says "Z-Image-Turbo"
Photo2
Style:
\- 1970's Movie scene
\- Old and slightly blurry
\- Wide shot
\- Cinematic shot
\- Bright vivid colors
\- Front view
\- Depth of Field
Characters:
\- Amanda, a 25 yr old woman with blonde hair and white tank top. She has a white large hat and large sunglasses that sits on top of her head
\- Steve, a 30 yr old man wearing a blue buttoned shirt
Setting:
\- A highway in Texas filled with grass and trees
Scene:
\- Steve is driving the car a light blue convertible mercedes benz.
\- Amanda is in the passenger seat looking out the side with a huge smile
\- At the very top is a huge colorful title that says "Z-Image-Turbo"
Photo3
Style:
\- Magazine cover
\- Professionally shot
\- Ultra high quality
\- Ultra high resolution
\- Shot with DSLR
Characters:
\- Olivia, a 22 yr old young woman with pale skin, black hair, winged eyeliner, slim face, sharp chin wearing a buttoned blouse with blue, green, and red geometric pattern. She wears ripped skinny jeans. Her make up is professionally done for a magazine photo shoot.
Setting:
\- Studio with pink walls
Scene:
\- Olivia is sitting in a wooden stool looking at the viewer fiercely.
\- The top of the photo has a title saying "VOGUE" and at the bottom it says "z-image-turbo edition" below it it says "January 5, 2026"
Photo4
Style:
\- Movie scene
\- Professionally shot
\- Ultra high quality
\- Ultra high resolution
\- Cinematic shot
\- From low angle
Characters:
\- Olivia, a 22 yr old young woman with pale skin, black hair, winged eyeliner, slim face, sharp chin wearing a buttoned blouse with blue, green, and red geometric pattern. She wears ripped skinny jeans.
Setting:
\- outdoors with blue sky
Scene:
\- Olivia is standing with one hand shielding her eyes from the bright sunlight.
\- A bright blue sky with a few clouds are in the background
\- The title of the movie is a stylized font saying "Z-Image-Turbo"
