alibaba/z-image/turbo/inpaint/lora
Z Image Turbo Inpaint by Alibaba - AI-powered image editing, style transfer, and transformation. Edit photos with natural language instructions, remove backgrounds, change styles, and enhance images effortlessly.
PNG, JPEG, WebP, or GIF · 20 MiB maximum
PNG, JPEG, WebP, or GIF · 20 MiB maximum
Example output — click Run to generate your own
Send your first request
OpenAI-compatible endpoint with unified authentication and usage tracking.
https://api.sandbase.ai/v1/runalibaba/z-image/turbo/inpaint/loraInput Schema
11 parameters · 2 required · 9 optional
| Parameter | Type | Required | Description |
|---|---|---|---|
image | string | Required | URL of Image for Inpaint generation. |
prompt | string | Required | The prompt to generate an image from. |
mask | string | Optional | URL of Mask for Inpaint generation. |
seed | integer | Optional | The same seed and the same prompt given to the same version of the model will output the same image every time. |
loras | object[] | Optional | List of LoRA weights to apply (maximum 3). · Default: [] |
strength | number | Optional | The strength of the inpaint conditioning. · Max: 1 · Default: 1 |
control_end | number | Optional | The end of the controlnet conditioning. · Min: 0 · Max: 1 · Default: 0.8 |
aspect_ratio | string | Optional | The aspect ratio of the generated image. · Options: 21:9, 16:9, 3:2, 4:3, 5:4, 1:1, 4:5, 3:4, 2:3, 9:16 21:916:93:24:35:41:14:53:42:39:16 |
control_scale | number | Optional | The scale of the controlnet conditioning. · Min: 0 · Max: 1 · Default: 0.75 |
control_start | number | Optional | The start of the controlnet conditioning. · Min: 0 · Max: 1 · Default: 0 |
output_format | string | Optional | The format of the generated image. · Options: jpeg, png · Default: "png" jpegpng |
Output Schema
| Field | Type | Description |
|---|---|---|
id | string | Unique identifier for the generation task |
status | string | Task status: pending, running, completed, failed, timeout |
model | string | Model used for the generation |
outputs | array | Array of output items |
outputs[].url | string | URL of the generated artifact |
outputs[].content_type | string | MIME type (e.g. image/png, video/mp4) |
error | object | null | Error details if failed, null on success |
error.type | string | Machine-readable error type code |
error.message | string | Human-readable error description |
Async Workflow
This model uses asynchronous execution. Submit a request and poll for the result.
- Submit — POST to /v1/run, receive an
id - Poll — GET /v1/run/{id} until status is
completed,failed, ortimeout - Retrieve — Read
outputsfrom the completed response
Code Examples
Ready-to-run snippets
# Step 1: Submit
curl -X POST https://api.sandbase.ai/v1/run \
-H "Content-Type: application/json" \
-H "Authorization: Bearer YOUR_API_KEY" \
-d '{
"model": "alibaba/z-image/turbo/inpaint/lora",
"mask": "https://static.sandbase.ai/examples/alibaba/z-image/turbo/inpaint/lora/input_mask_1.jpg",
"image": "https://static.sandbase.ai/examples/alibaba/z-image/turbo/inpaint/lora/input_image_0.jpg",
"loras": [],
"prompt": "A young Asian woman with long, vibrant purple hair stands on a sunlit sandy beach, posing confidently with her left hand resting on her hip. She gazes directly at the camera with a neutral expression. A sleek black ribbon bow is tied neatly on the right side of her head, just above her ear. She wears a flowing white cotton dress with a fitted bodice and a flared skirt that reaches mid-calf, slightly lifted by a gentle sea breeze. The beach behind her features fine, pale golden sand with subtle footprints, leading to calm turquoise waves under a clear blue sky with soft, wispy clouds. The lighting is natural daylight, casting soft shadows to her left, indicating late afternoon sun. The horizon line is visible in the background, with a faint silhouette of distant dunes. Her skin tone is fair with a natural glow, and her facial features are delicately defined. The composition is centered on her figure, framed from mid-thigh up, with shallow depth of field blurring the distant waves slightly.",
"strength": 1,
"control_end": 0.8,
"control_scale": 0.75,
"control_start": 0,
"output_format": "png"
}'
# Step 2: Poll result (replace <id>)
curl https://api.sandbase.ai/v1/run/<id> \
-H "Authorization: Bearer YOUR_API_KEY"API README
Z-Image Turbo Inpaint LoRA
Z-Image Turbo Inpaint LoRA Z-Image Turbo Inpaint LoRA reconstructs selected regions with the fast Z-Image Turbo model. The attached low-rank adaptation introduces a specialized aesthetic, subject, or production concept without replacing the underlying image model.
Mask only the intended area and describe the content, perspective, material, and lighting that should fill it. Include the adaptation’s learned concept naturally inside the brief so it interacts with the complete composition.
Highlights
- Z-Image Turbo Inpaint LoRA Masked content reconstruction. Generates new content only inside selected image regions.
- Z-Image Turbo Inpaint LoRA Boundary-consistent blending. Matches texture, perspective, color, and lighting around the mask.
- Z-Image Turbo Inpaint LoRA Prompt-directed replacement. Adds, removes, or transforms localized objects through language.
- Z-Image Turbo Inpaint LoRA Outside-mask preservation. Retains the source image beyond the requested edit area. A LoRA adds a learned style, subject, or domain concept to this behavior.
Pricing
| Billing unit | Price |
|---|---|
| Per request | $0.0115 |
When to Use
| ✅ Good fit | ❌ Consider alternatives |
|---|---|
| A selected time region in existing audio needs regeneration | The entire track should be recomposed |
| A managed asynchronous result is suitable for the production pipeline | A synchronous, interactive editor is essential |
| The documented controls cover the required duration, framing, or format | The project needs controls outside this endpoint's schema |
| Creative iteration benefits from a repeatable request structure | Exact deterministic pixels, frames, geometry, or samples are mandatory |
| A finished downloadable media asset is the desired deliverable | Editable source layers or a native project file are required |
Prompt Guide
For image-conditioned generation, state the intended result first, then add the subject or source treatment, progression, style, and delivery constraints. Keep one creative variable per phrase, use the documented field names for controls, and change one setting at a time when comparing results.
{
"aspect_ratio": "21:9",
"image": "https://example.com/start-frame.png",
"output_format": "png",
"prompt": "A young Asian woman with long, vibrant purple hair stands on a sunlit sandy beach, posing confidently with her left hand resting on her hip. She gazes directly at the camera with a neutral expression. A sleek black ribbon bow is tied neatly on the right side of her head, just above her ear. She wears a flowing white cotton dress with a fitted bodice and a flared skirt that reaches mid-calf, slightly lifted by a gentle sea breeze. The beach behind her features fine, pale golden sand with subtle footprints, leading to calm turquoise waves under a clear blue sky with soft, wispy clouds. The lighting is natural daylight, casting soft shadows to her left, indicating late afternoon sun. The horizon line is visible in the background, with a faint silhouette of distant dunes. Her skin tone is fair with a natural glow, and her facial features are delicately defined. The composition is centered on her figure, framed from mid-thigh up, with shallow depth of field blurring the distant waves slightly."
}
Technical Specs
| Spec | Value |
|---|---|
| Model ID | alibaba/z-image/turbo/inpaint/lora |
| Inputs | aspect_ratio, control_end, control_scale, control_start, image, loras, mask, output_format, prompt, seed, strength |
| Required inputs | prompt, image |
| Output fields | content_type, url |
| Execution | Async (submit, then poll for result) |
| Aspect Ratio | 21:9 / 16:9 / 3:2 / 4:3 / 5:4 / 1:1 / 4:5 / 3:4 / 2:3 / 9:16 |
| Output Format | jpeg / png |
Related Models
alibaba/z-image/turbo/inpaint— Compare this concrete local family route.alibaba/z-image/turbo— Compare this concrete local family route.alibaba/z-image/turbo/edit— Compare this concrete local family route.

