How good is Z-Image at rendering text in images?

In-image text is the headline strength of Z-Image. It renders legible English and Chinese — including small fonts and complex layouts — where generic image models typically produce garbled letters. Put the text you want in your prompt (in quotes) for posters, banners, and social graphics.

Does Z-Image support image editing?

No — Z-Image is text-to-image only. There is no image input. For editing an existing image, use a model that accepts an image_url. Z-Image focuses on fast, low-cost generation from a text prompt.

How does Z-Image pricing work?

Credit-based: 1 credit equals $0.001 USD. Z-Image charges a flat rate per generated image, independent of aspect ratio, and each request produces one image — it is one of the lowest per-image rates on the platform. Failed and rejected requests are refunded automatically. See the pricing card above for the current rate.

What aspect ratios does Z-Image support?

Z-Image accepts 1:1, 4:3, 3:4, 16:9, and 9:16, with a default of 1:1. There is no width/height or resolution control — pick the ratio and Z-Image renders accordingly.

Does Z-Image produce more than one image per request?

No — Z-Image returns exactly one image per request. To produce a set, send multiple requests. Each is billed as one image, so cost scales linearly and predictably.

Yes. You submit a Z-Image task and receive a task_id, then poll /api/v1/tasks/:id until it is completed. This is the same task flow as every other reApi model, and the API returns an image URL you should mirror to your own storage if you need it long term.

Where is the full API reference for Z-Image?

The complete parameter list, validation rules, error codes, and runnable cURL / Python / Node / Go examples for Z-Image live at /docs/z-image.

Z-Image — Fast, Low-Cost Image Generation with Bilingual Text

Z-Image is Alibaba's Tongyi Z-Image-Turbo: a fast, photorealistic text-to-image model that actually renders legible English and Chinese text — posters, banners, and social graphics included. It is one of the lowest-cost models on the platform, and you call Z-Image through reApi's OpenAI-compatible API.

Input

Prompt*

A realistic, high-resolution night photograph mimicking the aesthetic of flash photography, featuring a stylish young couple posing closely together on an elegant stone balcony with a classic white balustrade. The young woman, positioned on the left with her body angled slightly away, has long, voluminous honey-blonde wavy hair cascading down her back and is wearing a black spaghetti-strap mini dress featuring intricate lace detailing on the bodice and a tiered, ruffled skirt; she looks back over her shoulder at the camera with a soft, pleasant smile while resting one hand on the man's shoulder. Beside her stands a handsome young man with tan skin, dark curly hair styled with a fade, and a confident expression, dressed in a crisp black button-down shirt tucked into black dress pants with a visible leather belt, his arm wrapped affectionately around her waist. The setting is nighttime with a pitch-black sky and distant, out-of-focus city lights visible in the background, contrasting with the bright, direct lighting on the subjects and the white architectural column on the right, capturing a candid, trendy "date night" or homecoming social media vibe.

Up to 1000 characters · required

Aspect ratio

1:1 / 4:3 / 3:4 / 16:9 / 9:16 · default 1:1

Content filter

On in the playground; configurable via the API (API default: off).

Result

Try one of these prompts

What you can build with this model

Real-world workflows and production use cases you can build and ship with this model.

Z-Image social-media graphic with bilingual text

Social graphics with real text

Most image models turn captions into gibberish. Z-Image renders legible English and Chinese type, so OOTD collages, promo graphics, and social cards come back with the words spelled right and laid out where you asked. Write the copy in your prompt, set the aspect ratio, and it composes art and text together in one fast, low-cost generation.

Open the playground

Photorealistic people and fashion

It produces lifelike portraits and outfits with refined lighting and clean skin texture — strong enough for lookbooks, e-commerce, and street-style shots. At a compact 6B scale it stays fast and cheap, so you can iterate on poses, styling, and scenes without burning your budget on a single hero image.

Z-Image magazine cover with editorial typography

Magazine covers and editorial layouts

It is built for images where the typography matters as much as the photo. Generate magazine covers, banners, and posters with crisp English or Chinese headlines and balanced composition. Because the text stays sharp and on-layout, the output is usable as-is instead of needing a second pass in a design tool.

Pricing

Credit-based — 1 credit = $0.001 USD. Pay only for completed generations.

Category	Unit	Price
z-image
Per image	1 image	$0.005 5 credits

Why reAPI

Text that's actually legible

The reason teams reach for Z-Image is the in-image text. It renders accurate English and Chinese — small fonts and complex layouts included — where generic models smear letters. If your output has to carry words, it keeps them readable, through the same request to /api/v1/images/generations.

Fast and very cheap

Z-Image-Turbo runs in minimal sampling steps, so generations come back quickly. 1 credit equals $0.001 USD and it is one of the lowest per-image rates on the platform — flat per image, so cost is a single multiply, ideal for high-volume social and e-commerce work.

Async-first, OpenAI-compatible

Submit a Z-Image task, get a task_id back, poll until it completes. The JSON matches the OpenAI image-generations contract, so adding it is a model-id change, not a new integration. Failed and rejected requests are refunded automatically.

Z-Image vs Nano Banana

Both generate images from one API. Z-Image leans into legible bilingual text, photorealism, and a very low flat per-image price. Here is how the two compare on publicly documented behavior.

Capability

Z-Image on reAPI

Nano Banana

In-image text

A core strength — accurate English and Chinese type, even small fonts and complex layouts.

Renders text, with quality that varies by layout complexity.

Photorealism

Lifelike portraits and scenes from a compact 6B Tongyi model.

Strong photorealistic output.

Modality

Text-to-image only — fast generation from a prompt.

Generation plus prompt-driven editing.

Speed

Turbo: minimal sampling steps, low latency.

Fast, varies by request.

Pricing

Flat per image at one of the lowest rates on the platform.

Per-image pricing, typically higher.

Aspect ratios

1:1, 4:3, 3:4, 16:9, 9:16.

A set of common aspect ratios.

Comparison reflects publicly documented behavior at the time of writing. Model behavior and pricing can change; check the pricing card above and the API docs for current values.

Integrate Z-Image in three steps

step 01
Create an API key
Sign up at reApi and grab an API key. Free signup credits cover a lot of Z-Image generations — it is one of the cheapest models on the platform.
Open
step 02
Submit a generation
POST to /api/v1/images/generations with model: z-image, a prompt, and an optional aspect_ratio. Z-Image returns a task_id immediately.
Open
step 03
Poll the result
GET /api/v1/tasks/:id until status is completed. The Z-Image response carries the image URL; mirror it to your own storage if you need it long term, since generated links expire.
Open

Frequently asked questions

Common questions about this model.

Z-Image is Alibaba's Tongyi Z-Image-Turbo, a fast text-to-image model known for accurate bilingual (English and Chinese) text rendering and photorealistic output at very low cost. On reApi, Z-Image ships on the OpenAI-compatible /api/v1/images/generations endpoint under the model id z-image.

Related models

Explore more models in the same category.

View all models

Image

Alibaba

Qwen Image 2

Qwen Image 2 — Alibaba's unified text-to-image and editing model with strong in-image text rendering, on one OpenAI-compatible endpoint.

From $0.031 per image

Image

Alibaba

Wan 2.7 Image

Wan 2.7 Image — Alibaba Wan generation + editing, multi-image reference, and up to 4K, on one OpenAI-compatible endpoint.

From $0.032 per image

Image

Google

Gemini 2.5 Flash Image Preview

Nano banana — fast Gemini image generation and editing, up to 14 reference images at 1K.

From $0.015 per image

Image

ByteDance

Seedream 5.0 Lite

Seedream 5.0 Lite — Doubao image generation at 2K / 3K with image-to-image and batch output.

From $0.031 per image

View all models

start building

Ready to ship?

Try it in the playground or grab an API key to integrate now.

Try Z-Image View API Docs

Z-Image — Fast, Low-Cost Image Generation with Bilingual Text

What you can build with this model

Real-world workflows and production use cases you can build and ship with this model.