Skip to main content
Kyeo AI
AI image generation model
Tongyi-MAI
Z-Image

Z-Image

Kyeo AI's Z-Image workflow is a streamlined text-to-image option. The public source service charges a raw price of 0.8 credits per image, and the local integer-credit wallet rounds it up to 1 credit; prompts can contain up to 1,000 characters, and 5 aspect ratios are available. The workflow does not accept image input or provide editing.

Use this modelView credit pricingBack to Model Hub
Capability
Text-to-image
Credit cost
1 credit per request
Prompt limit
1,000 characters
Upload limit
Not applicable (no image input)
30-second overview
  • The Kyeo AI workflow is text-to-image only, with no image input or editing mode.
  • The interface provides a prompt and `aspect_ratio`, with 5 common ratio choices.
  • Prompts can contain up to 1,000 characters, so the main requirements need to be clear.
  • The public source service charges 0.8 credits per image; Kyeo rounds the total up, so one image uses 1 credit.

Popular searches: How do I use Z-Image? · What does Z-Image cost? · Does Z-Image cost 1 credit?

Examples

Visual concept

Visual concept

Aspect Ratio: 9:16 Vertical Duration: 15 Seconds Style: Premium street-fashion commercial, cinematic lifestyle advertising, realistic camera physics, luxury tech campaign, viral social-media edit. IMPORTANT: Treat each storyboard panel as an individual full-screen shot. Read left-to-right, top-to-bottom. Do NOT animate the collage. Every frame becomes a separate cinematic scene. Use the exact white-and-green wireless headphones from the reference images. Preserve ear-cup shape, hinge design, logo placement, premium matte finish, soft leather cushions, realistic materials, and proportions throughout all shots. ### Shot 01 (0.0–1.2s) Wide cinematic establishing shot. Young woman in oversized red shirt and black outfit walks through a red-brick urban alley. Headphones hang casually from her hand. Slow handheld push forward. Warm late-afternoon light. Fashion-commercial atmosphere. ### Shot 02 (1.2–2.2s) Macro product detail. Hand raises the white-and-green headphones. Shallow depth of field. Premium texture visible. Soft city bokeh. Gentle camera drift. ### Shot 03 (2.2–3.3s) Medium shot. She places headphones over her ears. Hair moves slightly in the wind. Match cut synchronized to soundtrack beat. Smooth handheld realism. ### Shot 04 (3.3–4.4s) Wide tracking shot. She walks deeper into the alley. Camera follows beside her. Natural parallax from buildings. Urban confidence. ### Shot 05 (4.4–5.5s) Extreme close-up of ear cup. Green accents catch reflected light. Premium material detail. Cinematic rack focus. ### Shot 06 (5.5–6.7s) Medium close-up profile. Eyes closed. Calm expression. Noise-canceling serenity. Slow push-in. Background softly blurred. ### Shot 07 (6.7–7.9s) Low-angle wide shot crossing a city intersection. Golden-hour reflections. Confident stride. Headphones centered naturally within frame. ### Shot 08 (7.9–9.1s) Tight side profile. Hair texture and skin detail contrasted against smooth headphone surface. Passing cars create subtle motion streaks. ### Shot 09 (9.1–10.3s) Medium portrait. She lifts one ear cup slightly. Small smile. Connecting with the city atmosphere. Gentle camera sway. ### Shot 10 (10.3–11.5s) Wide follow shot from behind. Walking toward a graffiti-covered wall. Red-and-black environment complements the green-and-white product. ### Shot 11 (11.5–12.8s) Macro hero product shot. Headphones rotate slowly in her hand. Dynamic reflections reveal premium craftsmanship. Sharp logo visibility. ### Shot 12 (12.8–15.0s) Hero ending. Woman seated on urban staircase. Headphones worn confidently. Soft backlight creates halo around hair. Slight breeze. Slow cinematic push-in. Peaceful yet powerful expression. Product remains crisp, dominant, and aspirational. **Camera Language** * Realistic handheld movement * Slow push-ins * Natural parallax * Motivated pans and tilts * Selective focus transitions * High-end commercial cinematography * No floating drone moves * No artificial zoom effects * No CGI camera tricks **Transitions** * Hard cuts on beat * Match cuts using hand movement * Match cuts using head turns * Momentum-driven edits * No wipes * No dissolves * No graphic overlays **Sound Design** * Distant city traffic * Soft wind * Footsteps on pavement * Clothing movement * Headphone handling sounds * Subtle headphone power-click effects * Deep ambient bass pulse * Modern cinematic electronic beat * Emotional luxury-tech atmosphere **Final Frame** Hold for 0.5 seconds on the seated hero shot. White-and-green headphones remain perfectly visible, logo clear, premium lighting, cinematic depth, luxury advertising finish, viral social-media campaign quality, ultra-realistic 4K HDR.

Visual concept

Visual concept

Ultra-realistic luxury product photography of a white creamy glass reed diffuser bottle labeled “WHITE OUDH – Private Collection – RITUALS” in elegant gold serif typography. The bottle has a curved rectangular shape with soft edges, a glossy reflective surface, and a gold wax-seal style emblem centered near the top. Thin black diffuser reeds extend upward from the bottle neck. The bottle is covered in realistic water AND ICE droplets and condensation, as if freshly misted after rain. It is placed among deep black roses and dark burgundy petals with visible water droplets, surrounded by rich green leaves. The scene is moody and dramatic with a dark botanical background, softly blurred (shallow depth of field, bokeh effect). Lighting is cinematic and low-key, with soft diffused highlights reflecting off the wet glass surface. Cool-toned lighting (deep greens and midnight blues) enhances the dark, luxurious atmosphere. The gold emblem and lettering subtly catch the light, creating a premium, high-end aesthetic. Composition: vertical portrait orientation (4:5 ratio), centered product placement, slightly angled bottle resting naturally among roses. Foreground roses partially frame the bottom of the bottle. Background is dark foliage fading into shadow. Texture emphasis: hyper-detailed water droplets, glossy glass reflections, velvety rose petals, matte black reeds. Style: high-end commercial fragrance advertising, macro lens, 85mm, f/2.0, shallow depth of field, photorealistic, 8K resolution, sharp focus on bottle, dramatic contrast, rich blacks, luxury perfume campaign aesthetic.

Visual concept

Visual concept

[BRAND NAME] + [COLOR] Act as a Senior CGI Artist and Brand Identity Director specializing in premium logo materializations. Your reference aesthetic: a logo mark that appears to have been pushed outward from behind a metallic or matte surface — like a relief stamp pressed from the reverse side of a metal sheet, a hallmark embossed on luxury packaging, or a raised seal on official documents. The logo does not exist as a separate standing object placed on a surface. It exists as a raised relief, a bulge, an outward protrusion that is part of the surface itself. BRAND INTELLIGENCE SYSTEM Before executing any phase, resolve these parameters: (1) LOGO GEOMETRY — identify [BRAND NAME]'s primary icon mark in its simplest most reduced form — the mark that works at any scale, (2) COLOR PALETTE — based on [COLOR], build the full tonal system from the lightest specular highlight to the deepest shadow tone, determine if the overall mood reads warm, cool, or neutral, (3) SURFACE MATERIAL — based on [COLOR] determine the surface character: silver or grey becomes brushed or circular-grain steel, gold or champagne becomes warm brushed metal with radial grain, black becomes anodized aluminum or matte carbon fiber, white becomes matte ceramic or chalk plaster, any saturated hue becomes anodized colored metal holding the hue in lit areas and desaturating toward black in deepest shadows, (4) LIGHT DIRECTION — determine the most flattering single light angle to reveal the emboss topography — where highlights land on the raised ridge peaks and where shadows fall on the descending walls. PHASE 1: SURFACE & ATMOSPHERE The entire image is one single continuous material surface — not a background with an object placed on it, but one unified physical plane filling the entire canvas. This surface is made of the SURFACE MATERIAL resolved above. The surface has a subtle radial gradient in its lighting — brighter near the center where the embossed logo sits, gradually darkening toward all four edges as the light falls off naturally. If the material is brushed metal, the grain direction is radial or concentric — circular polishing marks emanating from the center, catching the light differently at each angle. Apply fine uniform grain texture across the entire surface at ISO 800 equivalent — this gives tactility, depth, and prevents any digital flatness. The surface must feel physical and touchable — like a metal plate, a luxury tin lid, or a premium embossed card stock you could run your finger across and feel the texture. PHASE 2: THE EMBOSS — CRITICAL The [BRAND NAME] logo mark is rendered as a full bas-relief — think of a coin, a commemorative medal, or a luxury brand stamp where the ENTIRE logo shape rises as one unified solid mass from the surface. Not just the outline. Not just the stroke path. The complete filled silhouette of the logo pushes outward from the surface as a single continuous raised form — like the head on a coin pressed from a die. The interior of the logo is NOT hollow or open. The entire logo shape — every filled area, every curve, every solid region — rises together as one piece, at a uniform height above the surrounding flat surface. The top face of this raised form is slightly convex — gently domed, not perfectly flat — so it catches the key light across its full width and creates a smooth highlight gradient from the lit side to the shadow side. The transition from the flat surrounding surface up to the raised logo form follows a smooth beveled wall — curved, not sharp — like the edge of a coin. This beveled wall catches the key light on the lit side and falls into shadow on the opposite side, creating the dimensional separation between the raised form and the flat background. Any negative spaces WITHIN the logo — such as the bite in the Apple, the interior counters of letterforms — are recessed BACK to the level of the flat surface or slightly below, creating clean negative voids within the overall raised form. These recessed areas have their own small beveled walls descending inward, catching light in reverse. The overall raised height must feel substantial — like a thick coin or a heavy embossed seal — not a shallow bump. The logo geometry must be completely faithful to [BRAND NAME]'s actual mark in correct proportions and correct silhouette. PHASE 3: LIGHTING One primary soft area light from upper-left at approximately 10 to 11 o'clock position — broad and diffused, wrapping around the curved walls of the raised ridge to create the characteristic highlight peak at the top and gradual shadow descent on the far wall. Color temperature is determined by [COLOR]: cool white light for silver, grey, and blue tones — warm amber light for gold, copper, and earth tones — neutral daylight for white and black. One very subtle fill light from lower-right at 10 to 15 percent intensity of the key — just enough to retain material detail in the deepest shadow areas of the ridge walls without lifting them enough to flatten the form. No hard shadows anywhere on the surface. No spotlight pools. No rim light. The lighting exists entirely to reveal the three-dimensional topography of the emboss — every curve, every height transition, every subtle variation in the surface. PHASE 4: TYPOGRAPHY In the lower portion of the canvas, below the embossed logo, on the same continuous surface — a minimal typographic lockup. The typography is either very subtly embossed in the same material as the surface using the same raised-ridge technique at much smaller scale, or rendered as clean flat text in the lightest tone of [COLOR] sitting just at the surface level. Element 1: [BRAND NAME]'s flat logo mark at small scale — approximately 3 to 4 percent of canvas width — as a simple flat mark, not embossed. Element 2: [BRAND NAME] wordmark in clean geometric sans-serif spaced capitals directly below the icon mark. Element 3: one short italic or light-weight descriptor line significantly smaller than the wordmark — autonomously generate a relevant subtitle for [BRAND NAME] such as a founding year, a product category, a brand division, or a core tagline. All three elements stacked vertically, centered, with generous spacing. No decorative elements. No lines or rules. Pure typographic restraint. TECH SPECS The single most important instruction in this entire prompt: the logo must read as PUSHED OUTWARD FROM the surface, rising toward the viewer, proud and raised. If the result looks like a 3D logo object floating above a background it is wrong. If the result looks like a depression or hole cut into the surface it is wrong. The correct result looks like you are holding a sheet of metal that someone pressed a die stamp against from the back — and you are looking at the front where the relief has formed. Ray Tracing enabled for accurate self-shadowing — the raised ridge must cast a real shadow onto the flat surrounding surface. No Depth of Field — the entire surface is in perfect sharp focus from edge to edge. Tone mapping: preserve highlight detail — the brightest specular on the ridge peak should be brilliant and tight but never blown out or featureless. Film grain applied uniformly across the full final render. No chromatic aberration. No lens distortion. No added vignette beyond what the surface lighting gradient naturally creates. Anti-aliasing maximum — the ridge edges must be perfectly clean. Mood: a maker's hallmark, a foundry seal, a luxury embossed cover — the quiet confidence of a brand that lets its materiality speak.

Visual concept

Visual concept

Dreamy sunlit portrait of a young woman with long soft light-brown hair, delicate facial features, fair luminous skin, natural rosy lips, and expressive hazel-green eyes. Wearing an oversized cream-white knitted sweater, sitting beside a large window with warm golden morning sunlight streaming through. Soft shadows from the window frame fall across her face and clothing, creating a cozy and intimate atmosphere. Gentle wind slightly tousling her hair, relaxed pose with one hand resting near her face, minimal makeup, natural beauty, Scandinavian-inspired aesthetic, soft pastel tones, shallow depth of field, creamy bokeh, cinematic lighting, ultra-realistic skin texture, editorial fashion photography, emotional storytelling, serene and elegant mood, photorealistic, masterpiece quality, 8K HDR, 85mm lens, f/1.4. Enhanced Cinematic Version: Golden-hour indoor portrait, beautiful young woman bathed in warm sunlight, sitting peacefully near a sunlit window, long flowing light-brown hair glowing with backlight, oversized cozy knit sweater, soft freckles, radiant skin, dreamy gaze toward the camera, delicate facial symmetry, warm ivory and beige tones, luxury lifestyle photography, airy atmosphere, natural light masterpiece, cinematic depth, soft highlights, premium fashion editorial, Vogue-style portrait, ultra-detailed eyes, realistic textures, high-end photography, shallow depth of field, warm glow, photorealistic, 8K. Negative Prompt: low quality, blurry, overexposed highlights, underexposed shadows, bad anatomy, distorted face, extra fingers, extra limbs, asymmetrical eyes, plastic skin, heavy makeup, oversaturated colors, noise, grain, watermark, text, logo, motion blur, CGI, cartoon, anime, unrealistic proportions, low resolution, artifacts. Style Keywords: golden hour, cozy aesthetic, soft sunlight, dreamy portrait, warm glow, cream sweater, editorial photography, Scandinavian minimalism, airy atmosphere, natural beauty, cinematic realism, luxury lifestyle

Visual concept

Visual concept

{ "model": "gemini-nano-banana-pro", "mode": "image", "aspect_ratio": "4:5", "quality": "ultra", "style": "cinematic_realism", "scene": { "setting": "busy neon-lit city street at night, wet road after rain, glowing signboards and traffic behind subject", "weather": "light drizzle just ended, reflective pavement, humid atmosphere", "background": "soft bokeh city lights, moving cars, faint pedestrian motion blur, tall buildings with colorful signage", "time": "night" }, "subject": { "type": "young woman", "pose": "standing in the middle of a crosswalk, holding a smartphone close to camera", "expression": "natural warm smile, candid and joyful", "hair": "slightly messy hair blown across face by wind", "wardrobe": "dark coat or blazer, minimal accessories", "props": ["smartphone"] }, "composition": { "framing": "close-up portrait with wide-angle distortion, subject centered, phone in foreground", "camera_distance": "very close, intimate street-photo feel", "perspective": "wide lens perspective, slight fisheye look, background stretching outward", "depth_of_field": "shallow to medium, subject readable but background lights creamy", "foreground_detail": "phone and hands slightly soft due to motion and lens character" }, "camera": { "lens": "24mm wide-angle (slight fisheye character)", "aperture": "f/1.8", "shutter_speed": "1/40", "iso": "1600", "focus": "face focus, slight falloff on phone" }, "lighting": { "key_light": "neon signage spill lighting face with mixed cyan/green and warm highlights", "fill": "streetlight bounce from wet ground", "mood": "dreamy urban nightlife, soft contrast, romantic realism" }, "color_grading": { "palette": "teal, cyan, green neon mixed with warm skin tones", "contrast": "soft contrast with lifted shadows", "highlights": "glowy bloom on neon signs", "film_emulation": "35mm film look" }, "texture_and_detail": { "film_grain": "strong visible grain, authentic", "bloom": "subtle halation on lights", "sharpness": "slightly imperfect, organic street photography", "skin_texture": "natural pores, no plastic smoothing" }, "motion": { "hair_movement": "wind-blown strands across face", "background_motion": "light streaking from cars and walking people", "camera_feel": "handheld candid shot" }, "negative_prompt": [ "plastic skin", "overly sharp face", "clean studio lighting", "anime", "cartoon", "hyper-symmetrical face", "over-saturated neon", "HDR look", "fake depth blur edges", "extra fingers", "distorted hands", "text artifacts", "logo distortions", "AI glossy skin", "perfectly clean background" ] }

Visual concept

Visual concept

Create a Cannes-level sports fashion advertising poster for an original performance sneaker brand named "AEROSTEP". Use ultra-realistic live-action commercial photography. Preserve the exact structural logic of the reference image: an extreme low-angle street-level shot on a sunny boardwalk or beachside promenade, one adult athletic woman dominating the frame through forced perspective as she bends down to tie her sneaker, with clear blue sky filling most of the background, palm trees and a very small number of casual pedestrians receding behind her, and the shoe action as the absolute visual hero. Style lock: premium sportswear campaign photography, street-fashion realism, ultra-low-angle lifestyle advertising, bright midday natural sunlight, documentary-meets-editorial commercial image, no illustration, no CGI look, no style drift. Port logic: make the brand super-symbol the moment of bending down to tie the sneaker from an exaggerated ground-up perspective. The identity of the campaign must be the oversized near-camera shoe, the athletic fold of the body, and the giant-on-the-boardwalk scale contrast. This tying gesture must feel instantly ownable, dynamic, and recognizable as the signature visual code of the brand. Transit logic: reduce all non-essential environmental information so the image reads more like a flagship sneaker poster. Keep only the minimum boardwalk context needed to support scale and realism: a cleaner horizon line, fewer pedestrians, softer storefront detail, and more open negative space in the sky. The final image must feel more controlled, more premium, and more product-led. Core composition: full-body female figure seen from a dramatic ground-up perspective, body bent forward in a compact athletic fold, one hand tying the sneaker lace near the camera, one leg strongly foreshortened and dominant in the foreground, the torso and head leaning diagonally downward, short blonde hair moving lightly in the breeze. The shoe closest to camera must be the main product focal point. Keep the open promenade street, a very small number of distant figures, and minimal palm-tree cues only as support. Product focus: make the hero product a premium silver-white performance sneaker with technical mesh, layered panel construction, subtle reflective accents, sculpted outsole, and realistic lace tension. The shoe must feel contemporary, stylish, and performance-oriented. Preserve the same tying-shoelace action and the same near-camera product emphasis. The sneaker remains the absolute visual hero. Wardrobe styling: an adult woman wearing a minimal retro-sport knit set in muted sage green and deep green, consisting of a fitted ribbed crop top and matching athletic briefs or sport bottoms, paired with scrunched white socks. Styling should feel clean, sunlit, youthful, and premium, with no extra accessories beyond the sportswear and shoes. Human realism: adult female only, natural anatomy, realistic athletic body proportions, believable perspective distortion caused only by the extreme low-angle lens, correct shoulders, arms, pelvis, thighs, knees, calves, ankles, feet, and hands. Real skin only with subtle pores, natural skin texture, realistic limb compression, soft body volume, slight natural imperfections, no mannequin skin, no plastic smoothing, no anatomical errors. Lighting: strong natural midday sunlight with clear directional shadows on the pavement, crisp highlight on skin and shoes, luminous blue-sky ambience, realistic street contrast, and no muddy blacks. Keep the scene hot, bright, and sharp, with clean shadow edges and premium commercial clarity. Environment: a bright coastal urban promenade with pale pavement, one or two palm trees, a few very distant pedestrians, and extremely simplified low storefront forms far in the background. Preserve the same sense of scale contrast between the giant near-camera athlete and the tiny background world, but keep the environment quieter and more minimal than the reference. No extra props, no vehicles, no visual clutter. Color strategy: dominant palette of sky blue, sunlit skin tones, silver-white sneaker highlights, muted sage green, deep green, and warm concrete neutrals. Keep the image clean, summery, sporty, and modern. Art direction goals: product-first hierarchy, brand super-symbol clarity, flagship restraint, bold forced-perspective impact, premium sneaker desirability, authentic street-sport energy, clean commercial finish, no text, no watermark, no logo clutter, no black blotches, no low-resolution details. Negative prompt: different camera angle, eye-level shot, missing low-angle perspective, distorted anatomy, oversized deformed limbs, extra fingers, broken hands, warped sneaker shape, unreadable laces, muddy pavement, cloudy sky, busy street, excessive crowd, distracting storefront detail, plastic skin, CGI look, illustration style, watermark, headline text, logo clutter, black blotches, low-resolution shoe details.

At a glance

Is this model right for you?

Choose Z-Image when you want to create a new image directly from text with a predictable cost. For existing-image edits, multiple references, or resolution tiers, compare a workflow that exposes those controls.

What it does best
On Kyeo AI, Z-Image is a streamlined text-to-image option. The interface does not expose image input, editing, references, or quality tiers. Choose it by the inputs and controls your task needs, rather than assuming the result is limited to a draft or guaranteed to be final-ready.
Who it is for
It suits prompt-driven covers, social images, poster concepts, and visual-style exploration, as well as individuals or teams that want to try several prompts at a fixed per-request cost.
Why use it on Kyeo AI
Kyeo AI makes the scope explicit: a prompt plus `aspect_ratio`, a 1,000-character prompt limit, 5 ratio choices, and a fixed 1-credit charge. You can confirm both the input boundary and the cost before generating.

Key facts

Quickly evaluate whether this model fits your business use case.

Category
AI image generation model
Vendor
Tongyi-MAI
Model family
Z-Image
Integration
Current site integration
Platform model ID
z-image
Runtime
Asynchronous task
Prompt limit
1,000 characters
Upload limit
Not applicable (no image input)

Other names

The same model may have different names in official documentation, community discussions, and API integrations. This section brings them together for you.

z-image
Z Image
ZImage

Common questions

These practical questions focus on the task, input conditions, and result requirements you should confirm before choosing.

How do I use Z-Image?
What does Z-Image cost?
Does Z-Image cost 1 credit?
What is the Z-Image prompt limit?
Does Z-Image support image-to-image generation?
Which aspect ratios does Z-Image support?
Which tasks is Z-Image suited to?
Z-Image vs Nano Banana
Z-Image vs GPT Image 1.5
Can Z-Image edit an uploaded image?

Decision guide

Still deciding? Start with these key decision points.

1
Use it for text-only image generation

Choose Z-Image when the task does not require an existing image, references, editing, or quality tiers. The small input set and fixed per-request cost make the choice straightforward.

2
Compare other models for edits or references

If the task requires editing an existing image, preserving a reference subject, or choosing a resolution or quality tier, compare Nano Banana, GPT Image 1.5, or Flux 2 Flex first.

3
Start with a clear prompt

Choose the target ratio, then write one prompt that covers the subject, scene, composition, and style. Refine it from the actual result, and switch models only when you need a control this workflow does not provide.

Popular comparisons

Compare common alternatives on the same task to clarify differences in inputs, controls, and cost.

How do Z-Image and Nano Banana differ in image input and price?
Which should I use for text-only generation: Z-Image or Nano Banana?
How do Z-Image and GPT Image 1.5 differ in editing controls?
When is GPT Image 1.5 a better fit than Z-Image?
How do Z-Image and Flux 2 Flex compare on resolution controls and price?
Which tasks need Flux 2 Flex rather than Z-Image?

Model comparison table

Not sure which model to choose? Compare the key differences between this model and its alternatives at a glance.

Role on Kyeo AI
Z-Image
Low-cost text-to-image generation
Nano Banana
Supports both text generation and image editing
GPT Image 1.5
Supports generation, editing, and quality tiers
Workflow
Z-Image
Text-to-image only
Nano Banana
Text-to-image and image editing
Flux 2 Flex
Text-to-image and image editing
Prompt limit
Z-Image
1,000 characters
Nano Banana
5,000 characters
GPT Image 1.5
No published character limit
Current price
Z-Image
Fixed at 1 credit per request
Nano Banana
Fixed at 3 credits per request
Flux 2 Flex
14–24 credits per request
When to use it
Z-Image
Creating a new image from text only
GPT Image 1.5
When image editing or quality tiers are required
Nano Banana
When generation and editing belong in one workflow

Real-world experience

Practical guidance based on public sources, current on-site limits, and representative tasks.

A small control set does not imply low output quality

Kyeo AI exposes only a prompt and aspect ratio. That describes the interface, not a draft-only quality tier. Judge whether a result is usable from the prompt and generated image, not from the number of controls.

Assess price and workflow separately

The fixed 1-credit price reduces the cost of each attempt, but it does not answer whether the workflow fits. Confirm that text-to-image is enough, then decide from the result whether to iterate or use a model with editing, references, or quality tiers.

Capabilities

Text-to-image only

The workflow accepts only a text prompt and no reference image. It is suited to creating a new image from scratch, not making local edits to existing material.

5 aspect ratios

Choose 1:1, 4:3, 3:4, 16:9, or 9:16 for square, landscape, or portrait composition. The interface does not provide a separate pixel-size or quality control.

Clear prompt and pricing limits

Prompts can contain up to 1,000 characters, and each request costs a fixed 1 credit. The price does not vary by aspect ratio, quality, or resolution in the current workflow.

Use cases

Concept sketches

Use a text prompt to test a subject, setting, composition, and visual style, then adjust the wording from the actual result.

Covers and social images

Generate article covers, short-video thumbnails, and social images in ratios such as 1:1, 16:9, or 9:16. Decide whether to publish the image only after reviewing the actual output.

Prompt and composition exploration

Compare themes, compositions, or style descriptions one request at a time with a predictable cost. The current page does not provide a batch-generation control.

Prompt tips

Structure the prompt by priority

Within the 1,000-character limit, state the subject, scene, composition, lighting, and style in that order. Put essential requirements first and avoid conflicting modifiers.

Choose the output ratio first

Match the aspect ratio to the intended placement before submitting. This interface does not provide an outpainting or post-generation reframing control.

Specify and verify visible text

If the image needs text, provide the exact wording, language, placement, and style.

Why choose it

A fixed 1-credit price makes the cost clear before generation.
The workflow requires only a prompt and aspect ratio.
5 ratios cover common square, landscape, and portrait compositions.
The 1,000-character allowance is sufficient for a well-structured scene description.

What to know first

The current workflow is text-to-image only; it cannot accept a reference or edit an existing image.
The 1,000-character prompt limit requires long or highly constrained briefs to be condensed.
The interface does not expose resolution, quality, seed, or negative-prompt controls.
Visible text, fine details, and subject consistency still require human review; a low price does not remove the need for quality checks.

Adjustable settings

Quickly evaluate whether this model fits your business use case.

Aspect ratio
Optional
Parameter type: Select
Default: 1:1
1:1
4:3
3:4
16:9
9:16

Credit cost

Public source 0.8/image; local 1 credit

The public source service currently charges a raw price of 0.8 credits per image. Kyeo preserves that decimal fact and applies `Math.ceil` to the request total, so one-image preauthorization and settlement use 1 credit.

Credit-saving tip

Before generating images or videos in bulk, run an A/B test with the same input on this model and similar alternatives. Confirm the result first to avoid wasting credits.

Frequently asked questions

Related models

Still deciding? Take a quick look at these similar alternatives.

Nano Banana

Nano Banana, also known as Gemini 2.5 Flash Image, provides text-to-image generation and natural-language editing on Kyeo AI. Requests without images use `Nano Banana`; requests with images use `Nano Banana`. Both paths cost 3 credits and accept prompts up to 5,000 characters.

AI image generator and editor
3 credits per request
View detailsUse now

GPT Image 1.5

GPT Image 1.5 on Kyeo supports text-to-image and image-to-image with three aspect ratios and medium/high quality. Image editing accepts up to 16 JPG, PNG, or WebP references at 10 MB each. Medium costs 4 credits and high costs 22 credits. GPT Image 1.5 is OpenAI's deprecated previous image model. Kyeo retains generation and editing access, but that availability does not make it OpenAI's current default choice.

AI image model
4–22 credits per request
View detailsUse now

Flux 2 Flex

Flux 2 Flex supports prompt-only generation and editing with up to eight images on Kyeo. A request without images uses Flux 2 Flex; an upload switches it to Flux 2 Flex. Seven fixed aspect ratios and 1K or 2K output are available for 14 or 24 credits per request.

AI image model
14–24 credits per request
View detailsUse now

Sources

This page is based on vendor materials, model-family documentation, and Kyeo AI integration settings.

Source note

Explore Z-Image on Kyeo AI: text-to-image, a 1,000-character prompt limit, 5 aspect ratios, an public source price of 0.8 credits per image, and a local total rounded up to 1 credit. The public source service currently charges a raw price of 0.8 credits per image. Kyeo preserves that decimal fact and applies `Math.ceil` to the request total, so one-image preauthorization and settlement use 1 credit.

Last updated: 2026-08-04
Tongyi-MAI / Z-Image GitHub
Official
Open original source
Z-Image Technical Report (arXiv)
Official
Open original source