Cookie Preferences
We use cookies to personalize content and analyze traffic. See our Cookie Policy.
No image yet. Submit the form to generate image.
Explore different use cases and parameter configurations
Transparent pricing with no hidden fees. Pay as you go.
| Rule & Modality | Channel | Credits | Price (USD) | Official / Reference Price | Daily Savings |
|---|---|---|---|---|---|
z-image-turbo imageqwen | Starter | 0.15per run | $0.001 | - | - |
Complete guide to using z-image
Integrate z-image/turbo through ApiPass to generate photorealistic images from natural-language prompts with low-latency asynchronous tasks. The model is designed for production workloads that need speed, throughput, realistic output, reliable text rendering, and affordable image generation.
Z-Image-Turbo is a 6 billion parameter text-to-image model from Tongyi-MAI, engineered for fast image generation with production-ready visual quality. It uses an efficient low-step generation design to produce full images quickly while maintaining strong photorealism, composition, and on-image text capability. Through ApiPass, z-image/turbo is exposed as an asynchronous REST API for text-to-image generation and optional single-reference-image guided generation. Each task produces one image, and completed results are returned as image URLs in resultJson.resultUrls.
Create high-quality photorealistic images from natural-language prompts. Z Image Turbo is well suited for cinematic scenes, product visuals, hero banners, UI imagery, thumbnails, social content, and rapid creative iteration.
Provide one publicly accessible reference image URL and use the prompt to describe the desired transformation. The strength parameter lets you control how strongly the output follows the prompt versus preserving the reference image.
Submit a generation task with POST /api/v1/jobs/createTask, save the returned taskId, then poll GET /api/v1/jobs/recordInfo every 3 to 5 seconds or provide callBackUrl for completion callbacks. On success, read the generated image from resultJson.resultUrls[0].
Z Image Turbo is optimized for low-latency image creation while preserving realistic lighting, texture, mood, and composition. It is a strong fit for interactive products, design tools, dashboards, and high-volume content backends.
The model understands English and Chinese prompts and can render multilingual text directly inside generated images, making it useful for posters, campaign visuals, UI mockups, social graphics, and cross-market creative workflows.
Use the size parameter as a WIDTH*HEIGHT pixel string such as 1024*1024. ApiPass passes pixel dimensions as strings, enabling square or rectangular output sizes while requiring the half-width asterisk format.
Set seed to -1 for random results or use a fixed integer to help reproduce similar outputs. This is useful for controlled experimentation, brand review, prompt iteration, and generating consistent creative variations.
Build fast prompt-to-image experiences for creative apps, chatbots, design assistants, and internal tools where users expect a quick click-to-image workflow.
Generate product photos, campaign concepts, hero banners, thumbnails, social media images, and catalog visuals at scale without slowing down production workflows.
Use a single reference image and a transformation prompt to create stylized variants, scene changes, visual refreshes, or controlled reinterpretations while tuning strength from subtle to dramatic.
Create posters, manga-style panels, promotional graphics, and UI mockups that include visible text in the image, with prompt-level control over style, atmosphere, and layout.
Use these implementation notes to integrate z-image/turbo reliably in production applications on ApiPass.
The Create Task API returning HTTP 200 with code 200 only means the task was accepted. It does not mean the image is complete. Save data.taskId and continue with polling or callbacks.
Poll /api/v1/jobs/recordInfo every 3 to 5 seconds. Continue polling for waiting, queuing, and generating. Stop immediately on success or fail. On success, read resultJson.resultUrls; on fail, read failCode and failMsg.
Only send the documented input fields: prompt, image, size, strength, seed, and output_format. The size value must be a WIDTH*HEIGHT pixel string such as 1024*1024, not 1024x1024 or an aspect-ratio label.
The source model card lists pricing from $0.005 per generated image. Actual charges may depend on ApiPass routing, account configuration, channel selection, and current billing rules, so check the live ApiPass price shown for your request.
ApiPass supports channel options including auto, starter, regular, and official. The default auto channel routes tasks based on available providers, pricing, and stability metrics.
Z Image Turbo is designed for efficient generation, making it practical for large batches such as catalog imagery, feed assets, auto-generated thumbnails, and repeated creative testing.
Start with a text prompt, choose a size such as 1024*1024, keep seed at -1 for random results, and select jpeg, png, or webp as the output format. Add a reference image URL only when you need guided generation.
Call POST https://api.apipass.dev/api/v1/jobs/createTask with model set to z-image/turbo and all model parameters inside input. Store the returned data.taskId for status checks.
Query GET /api/v1/jobs/recordInfo?taskId=<TASK_ID> every 3 to 5 seconds, or provide callBackUrl when creating the task. Stop when the task reaches success or fail.
When data.state is success, read the generated image from resultJson.resultUrls[0]. Each z-image/turbo task typically returns one generated image URL.
All APIs require authentication via Bearer Token.
Authorization: Bearer
Submit a new z-image/turbo asynchronous image generation task for text-to-image or single-reference-image guided generation.
The API accepts a JSON payload with the following structure:
1{
2 "model": "z-image/turbo",
3 "callBackUrl": "https://example.com/webhooks/apipass",
4 "channel": "auto",
5 "input": {
6 "prompt": "string",
7 "image": "string (optional)",
8 "size": "1024*1024",
9 "strength": 0.6,
10 "seed": -1,
11 "output_format": "jpeg"
12 },
13 "fallback_models": null
14}modelRequiredstringPublic model identifier. Must be exactly z-image/turbo, including capitalization and spelling.
"z-image/turbo"
inputRequiredobjectImage generation parameters. All model-specific fields must be placed inside this input object.
{"prompt":"A cinematic Shanghai street corner cafe after rain"}callBackUrlOptionalstring or nullPublicly accessible HTTP or HTTPS URL that receives a POST callback after the task succeeds or fails. If omitted or null, no callback will be sent. The field name must be written exactly as callBackUrl.
"https://example.com/webhooks/apipass"
channelOptionalstringRouting and billing channel. The default channel is auto. APIPASS automatically allocates tasks across available providers based on real-time pricing and stability metrics. starter is ultra-low-cost with limited quotas and weaker stability. regular is standard and much cheaper than official APIs with moderate stability. official uses the model native API with high stability, fast task execution, and official pricing.
The default channel is auto. APIPASS automatically allocates tasks across available providers based on real-time pricing and stability metrics; starter is ultra-low-cost with limited quotas and weaker stability; regular is standard and much cheaper than official APIs with moderate stability; official uses the model native API with high stability, fast task execution, and official pricing.
Available options:
auto
fallback_modelsOptionalarray<object> or nullAdvanced model fallback configuration. Each item must be an object containing model, channel, and the target model's complete input. The fallback input is not automatically merged with the primary input. Omit this field for normal z-image/turbo calls.
null
The input object contains the following parameters:
input.promptRequiredstringText description of the image to generate. For reference-image guided generation, this describes how the reference should be transformed.
Required. Trim leading and trailing whitespace before validation; the prompt must contain at least one non-whitespace character. When image is provided, use the prompt to describe the desired modification or target result.
"A cinematic Shanghai street corner cafe after rain, warm neon lights, realistic photography"
input.imageOptionalstringSingle reference image URL. Omit this field for text-to-image generation; provide it for reference-image guided generation.
Must be a publicly accessible HTTP or HTTPS image URL that the server can fetch directly. Do not pass local browser paths, URLs that require cookies or login state, or URLs that only work after temporary redirects. This field accepts one URL string, not an array. Common JPEG, PNG, and WebP image URLs are recommended.
"https://example.com/reference.png"
input.sizeOptionalstringOutput pixel size string. Defaults to 1024*1024. ApiPass receives this as a string and passes it as pixel dimensions.
Use WIDTHHEIGHT with a half-width asterisk, such as 10241024. Do not use 1024x1024 or aspect-ratio strings. The current public schema does not define a complete fixed enum; validate the format before submitting.
"1024*1024"
input.strengthOptionalnumberReference-image transformation strength. Valid range is 0 to 1 inclusive. Higher values usually produce results that differ more from the original reference image. Default: 0.6.
Only meaningful when input.image is provided. Use a slider from 0 to 1 with a recommended step of 0.01. 0 stays closest to the reference image, 0.3 is light modification, 0.6 is balanced, 0.8 follows the prompt more strongly, and 1 produces the largest change. These labels are UI guidance, not a strict mathematical guarantee.
0.6
input.seedOptionalintegerGeneration seed. Any integer is accepted; -1 means random seed. Default: -1.
Use -1 for a random seed. A fixed seed with the same parameters can help reproduce similar results, but pixel-level identical output is not guaranteed.
42
input.output_formatOptionalstringOutput image format. Default: jpeg.
Only use the listed values. Do not pass jpg, tiff, or other unsupported formats.
Available options:
"jpeg"
1curl -X POST "https://api.apipass.dev/api/v1/jobs/createTask" \
2 -H "Authorization: Bearer YOUR_API_KEY" \
3 -H "Content-Type: application/json" \
4 -d '{
5 "model": "z-image/turbo",
6 "callBackUrl": "https://example.com/webhooks/apipass",
7 "channel": "auto",
8 "input": {
9 "prompt": "A cinematic Shanghai street corner cafe after rain, warm neon lights, realistic photography, rich details",
10 "image": "https://example.com/reference.png",
11 "size": "1024*1024",
12 "strength": 0.6,
13 "seed": 42,
14 "output_format": "png"
15 }
16 }'1{
2 "code": 200,
3 "message": "success",
4 "data": {
5 "taskId": "task_555ad751dff448f8"
6 }
7}codeBusiness status code. 200 means the task was accepted successfully.
messageResponse message. For a successful task creation request, this is usually success.
data.taskIdApiPass task ID. Save this value and use it to query task status and results.
flux
flux
Generate photorealistic images with robust text
Starting from
0 credits
Nano Banana 2 Lite, aka Gemini 3.1 Flash-Lite Image, is a newly released image model from Google DeepMind built for fast 1K image generation, efficient creative iteration, prompt-based image editing, and responsive visual output in product-facing creative scenarios.
Starting from
0 credits
OpenAI
OpenAI
Generate intentionally designed visuals with less prompting using the latest OpenAI GPT Image 2 model.
Starting from
0 credits
Nano Banana now allows you to freely draw on edited images, bringing you a completely new experience. Simply click on the image to draw whatever you want. You can also add text and details, and Gemini will do the rest.
Starting from
0 credits
Fast image generation model built on Gemini 3.1 Flash Image. Supports text-to-image, image editing, multi-image fusion with up to 14 reference images, Google Search grounding, and outputs up to 4K resolution.
Starting from
0 credits
seedream
seedream
Starting from
0 credits
grok
grok
Grok Imagine is xAI’s AI video generation model for creating high-quality videos with realistic motion, native audio, and synchronized speech. Grok Imagine 1.5 introduces faster generation speeds, improved motion physics, and stronger visual consistency for modern AI video applications.
Starting from
0 credits
seedream
seedream
Starting from
0 credits
OpenAI
OpenAI
Bring your ideas to life with sharper, faster, more precise image generation using the GPT Image 2.5 API.
Starting from
0 credits
wan
wan
Generate images with cinematic precision through high-fidelity text-to-image synthesis and structural editing.
Starting from
0 credits
seedream
seedream
Gemini said Generate images with lightning-fast efficiency and high-fidelity detail through a streamlined, lightweight architecture optimized for instant creative expression.
Starting from
0 credits
qwen
qwen
Generate realistic 2K images with structured text
Starting from
0 credits
Luma
Luma
Starting from
0 credits
seedream
seedream
Starting from
0 credits