Use Z Image in OpenClaw.
Z Image is Alibaba's fast image generation model with bilingual text rendering in Chinese and English. OpenClaw agents call it through the same RunAPI endpoint and API key used for chat — no extra skills or provider accounts needed.
Use the RunAPI text_to_image endpoint to generate an image with Z Image.
Requirements:
- Use the RUNAPI_API_KEY environment variable for authentication.
- Send a POST request to https://runapi.ai/v1/text_to_image.
- Set model to "z-image".
- Write a descriptive prompt for the image you want to generate.
- The response returns a task_id. Poll the task status endpoint until the task completes, then retrieve the image URL from the output.
curl -X POST https://runapi.ai/v1/text_to_image \
-H "Authorization: Bearer $RUNAPI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "z-image",
"prompt": "A neon-lit Tokyo street at night with Japanese shop signs, cinematic lighting, 8K detail"
}'
{
"task_id": "tsk_abc123",
"status": "pending",
"model": "z-image"
}
Use Z Image in OpenClaw in three steps
Configure RunAPI
Set your RunAPI API key as an environment variable. If you already configured RunAPI for chat in OpenClaw, the same key works for image generation.
export RUNAPI_API_KEY=runapi_xxx
Call Z Image
Send a POST request to the text_to_image endpoint with model set to z-image and a text prompt. Z Image supports both Chinese and English prompts with accurate text rendering.
text_to_image
Get the result
The API returns a task_id immediately. Poll the task status endpoint until the status changes to completed, then retrieve the generated image URL from the response output.
task_id: tsk_abc123
Z Image API parameters
| Parameter | Type | Description |
|---|---|---|
model |
string |
Required. Use z-image. |
prompt |
string |
Text description. Supports Chinese and English. |
negative_prompt |
string |
Optional. What to avoid in the image. |
size |
string |
Optional. Output dimensions, e.g. 1024x1024. |
What is Z Image on OpenClaw?
Z Image is the speed-optimized image model that users call "amazing" for its sub-second generation and ultra-realistic output on consumer-grade hardware. It runs on Apache 2.0 open source, renders bilingual text in Chinese and English, and captures subtle facial details that other fast models miss. At around $0.004 per image through RunAPI, it is one of the cheapest options available. OpenClaw agents call it with the same API key used for chat.
Z Image use cases
Concept exploration and rapid prototyping
Generate dozens of visual concepts in seconds at fractions of a cent each, ideal for exploring directions before committing to a more expensive model for final output.
Bilingual marketing campaigns
Create marketing visuals with accurately rendered Chinese and English text in a single image, useful for cross-border e-commerce and multilingual social media.
Fashion and product photography mockups
Generate realistic product photography mockups with accurate material textures and lighting, fast enough to iterate on compositions in real time.
Z Image + OpenClaw questions
Yes. If you have RunAPI configured as a provider in OpenClaw, the same API key and endpoint handle Z Image requests. Send a POST to the text_to_image endpoint with model set to z-image.
No. The same RUNAPI_API_KEY you use for chat and other models also works for Z Image. All RunAPI models share one key and one credits balance.
Z Image is async. The API returns a task_id immediately. Poll the task status endpoint or use a webhook callback to know when the image is ready. RunAPI SDKs handle polling automatically.
Z Image uses pay-as-you-go pricing at around $0.004 per image, making it one of the cheapest options on RunAPI. Check the RunAPI pricing page for the current rate. There is no monthly subscription or minimum spend.
Z Image wins on speed and cost -- sub-second generation at a fraction of a cent per image. Flux 2 and Nano Banana produce higher-fidelity output but cost 5-10x more and take longer. Use Z Image for volume work and concept exploration, then switch to Flux 2 or Nano Banana for final selects.
Yes. Z Image renders bilingual text accurately in both Chinese and English within a single generated image. You can mix languages in the prompt and the model handles typography for both scripts, which is useful for cross-border marketing materials and multilingual product labels.
OpenClaw general setup
Not configured yet? Start with the RunAPI setup guide for OpenClaw.
OpenClaw setup guide →Try Z Image in OpenClaw today.
Get a free RunAPI key, paste the prompt into OpenClaw, and start generating images with bilingual text rendering.