by AlibabaImage

Z-Image

Photoreal images in about a second

5 aspect ratios

From 2 credits per image≈ $0.02

What it's like to use

Z-Image is a compact 6-billion-parameter image model from Alibaba's Tongyi team, tuned for photoreal results in about a second. Prompts of up to 1,000 characters, five aspect ratios and lossless PNG make it a fast way to try twenty directions before you commit.

Good for

Exploring a dozen compositions in a minute
Cinematic stills: noir streets, rainy crosswalks, aurora skies
Candid photos that read like a real phone snapshot

Real results from Z-Image

Every image and clip here was generated with Z-Image. Hit Recreate to open it in the studio with the same prompt and settings.

Black-and-white photo of a man in a trench coat and felt hat on wet cobblestones under an elevated railway, his breath fogging as a train passes overhead. — Create Image

A man in a charcoal overcoat holds his umbrella over a laughing woman in a camel coat on a rainy crosswalk at dusk. — Create Image

A hooded figure with a lantern stands on a stone bridge in a misty cavern before an enormous coiled dragon. — Create Image

A lone motorcyclist rides a cracked desert highway toward a towering black monolith in pale morning haze. — Create Image

A hiker in a red jacket stands small inside a vast blue glacier cave, sunlight pouring through a crack in the ice ceiling. — Create Image

A small red cabin with a lit window on a snowy lakeshore under green and violet aurora, two sled dogs resting beside it. — Create Image

A steaming bowl of ramen with a soft egg, charred pork and scallions on a dark wooden counter under a hanging bulb. — Create Image

A red fox curled asleep in fresh snow beside frozen reeds, a warm farmhouse window glowing in the blue dawn behind it. — Create Image

A breakdancer balances upside down on both hands on a tiled subway platform as commuters walk past and a train arrives. — Create Image

A steel chronograph with a plain grey dial on a block of frosted ice on black stone, lit by a cold blue backlight. — Create Image

What Z-Image can do

Each mode opens in the studio with Z-Image already picked. The price shows before you generate.

Create ImageTurn a written prompt into a finished image.From 2 creditsper image

How to prompt Z-Image

Keep the brief tight and concrete: spend the 1,000 characters on light, lens and texture.
For a candid look, ask for the flaws: fluorescent cast, motion blur, uneven focus.
Save the prompt and seed with any keeper; with the seed held, the same prompt lands very close to the earlier frame.

Settings worth knowing

Everything else has a sensible default, so you can type a prompt and go.

aspect_ratio1:1 · 4:3 · 3:4 · 16:9 · 9:16: The shape of the frame: wide, square or tall.
seed0–2147483647: Reuse the same number to get a repeatable result.
enable_prompt_expansion: Lets the model add detail to a short prompt.
output_formatpng: File type of the result.

How it compares

Same kind of job, different strengths and prices.

Z-Image Turbo LoRA runs the same family with your own LoRAs and adds image to image.

Z-ImageThis modelFrom 2 credits

Qwen Image 2.1 ProUp to six detailed 2K images from one promptFrom 6 creditsper image Nano Banana 2.1Google's newest Nano Banana, sharp text and 4K outputFrom 8 creditsper image Qwen Image 2.1 TurboFast 8-step renders, up to six per promptFrom 3 creditsper image Nano Banana ProPlans the layout first: dense text, diagrams and 4K editsFrom 17 creditsper image GPT Image 2.5 SunburstOpenAI's quality-first model, up to 4K with clean textFrom 1 creditper image Seedream 5.0 ProHigh-detail images and edits from up to ten referencesFrom 6 creditsper image

Questions about Z-Image

How much does Z-Image cost?

On PopNow, the cheapest run of Z-Image costs 2 credits ($0.02) per image. You pay with credits (1 credit = $0.01), the exact cost shows on the Generate button before you start, and failed generations are refunded automatically.

What can Z-Image do?

On PopNow, Z-Image works in these modes: Create Image. Pick one in the studio and add a prompt or an upload.

How should I write prompts for Z-Image?

Keep the brief tight and concrete: spend the 1,000 characters on light, lens and texture. For a candid look, ask for the flaws: fluorescent cast, motion blur, uneven focus. Save the prompt and seed with any keeper; with the seed held, the same prompt lands very close to the earlier frame.

Z-Image or Z-Image Turbo LoRA?

Z-Image Turbo LoRA runs the same family with your own LoRAs and adds image to image.

Do I need to install anything to use Z-Image?

No. Z-Image runs in your browser on PopNow: write a prompt or add a file, choose the settings and download the result. Sign up and you can generate straight away.

Building an app? See the Z-Image API →

Make something with Z-Image

Type an idea, pick your settings and see the price before you hit Generate.

Try Z-Image All models