alibaba/wan-2.7/image-edit
alibaba/wan-2.7/image-edit
WAN 2.7 Image Edit performs prompt-driven image editing with support for multiple-image references. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.
Examples
Parameters
| Name | Type | Default | Constraints | Description |
|---|---|---|---|---|
| prompt *Prompt | textarea | — | ≤ 5000 chars | The positive prompt for the generation. |
| images *Images | image_upload_group | — | image/* · 1–9 items | List of URLs of input images for editing (1-9 images). |
| sizeSize | text | — | ≤ 500 chars | The size of the generated image in pixels (width*height). Range: 512-4096 per dimension. Total pixels must be between 768*768 and 2048*2048. Aspect ratio must be between 1:8 and 8:1. |
API
Call this model through one unified REST API. Get a key on the API Keys page.
cURL
# 1) Submit — returns { "task_uuid": "..." }
curl -X POST "https://www.namifusion.com/api/v1/marketplace/run/alibaba/wan-2.7/image-edit" \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"input": {
"prompt": "A cinematic portrait, dramatic rim light, shallow depth of field",
"images": [
"https://example.com/input.jpg"
]
}
}'
# 2) Poll until status is "completed", then read the output URLs
curl "https://www.namifusion.com/api/v1/marketplace/run/tasks/TASK_UUID" \
-H "Authorization: Bearer YOUR_API_KEY"Python
import time, requests
API_KEY = "YOUR_API_KEY"
HEADERS = {"Authorization": f"Bearer {API_KEY}"}
# 1) Submit
resp = requests.post(
"https://www.namifusion.com/api/v1/marketplace/run/alibaba/wan-2.7/image-edit",
headers=HEADERS,
json={
"input": {
"prompt": "A cinematic portrait, dramatic rim light, shallow depth of field",
"images": [
"https://example.com/input.jpg"
]
}
},
)
resp.raise_for_status() # 401/402/429/5xx stop here instead of polling a bad task
task = resp.json()
# 2) Poll until a terminal state (completed / failed / cancelled).
# This model is allowed up to 300s server-side.
deadline = time.time() + 360
while task.get("status") not in ("completed", "failed", "cancelled"):
if time.time() > deadline:
raise TimeoutError(f"still {task.get('status')} — keep the task_uuid and poll later")
time.sleep(3)
poll = requests.get(f"https://www.namifusion.com/api/v1/marketplace/run/tasks/{task['task_uuid']}", headers=HEADERS)
poll.raise_for_status()
task = poll.json()
print(task["status"], task.get("output"))JavaScript
const API_KEY = "YOUR_API_KEY";
const HEADERS = { Authorization: `Bearer ${API_KEY}` };
// 1) Submit
const resp = await fetch("https://www.namifusion.com/api/v1/marketplace/run/alibaba/wan-2.7/image-edit", {
method: "POST",
headers: { ...HEADERS, "Content-Type": "application/json" },
body: JSON.stringify({
"input": {
"prompt": "A cinematic portrait, dramatic rim light, shallow depth of field",
"images": [
"https://example.com/input.jpg"
]
}
}),
});
if (!resp.ok) throw new Error(`submit failed: ${resp.status} ${await resp.text()}`);
let task = await resp.json();
// 2) Poll until a terminal state (completed / failed / cancelled).
// This model is allowed up to 300s server-side.
const deadline = Date.now() + 360 * 1000;
while (!["completed", "failed", "cancelled"].includes(task.status)) {
if (Date.now() > deadline) throw new Error(`still ${task.status} — keep the task_uuid and poll later`);
await new Promise((r) => setTimeout(r, 3000));
const poll = await fetch(`https://www.namifusion.com/api/v1/marketplace/run/tasks/${task.task_uuid}`, { headers: HEADERS });
if (!poll.ok) throw new Error(`poll failed: ${poll.status}`);
task = await poll.json();
}
console.log(task.status, task.output);Documentation
alibaba/wan-2.7/image-edit
Prompt-driven image editing with multi-image references, fast inference, and no cold starts.
alibaba/wan-2.7/image-edit is a prompt-based image editing model designed for targeted modifications to existing images. It preserves subject identity, composition, and scene structure while enabling fast creative iteration with multiple reference images, making it a strong fit for production workflows that prioritize quality, speed, and cost efficiency.
🚀 Key Features
- Prompt-driven editing: Describe what to change and what to preserve in natural language for precise, workflow-friendly image edits.
- Multi-image reference support: Accepts 1–9 input images for subject, style, background, or fusion-based editing tasks.
- Composition preservation: Optimized to maintain identity, pose, layout, and overall scene structure while applying localized changes.
- No cold starts: Ready for responsive production usage and rapid iteration without startup delays.
- Affordable per-run pricing: Low-cost execution makes it suitable for high-volume creative and commercial editing workloads.
🛠️ Technical Specifications
| Item | Details |
|---|---|
| Model | alibaba/wan-2.7/image-edit |
| Model Type | Prompt-driven image editing |
| Task Type | image-to-image |
| Architecture | Conditional image editing model with multi-reference guidance |
| Input | Prompt, 1–9 input images |
| Output | Edited image |
| Num Images | 1–9 |
| Resolution | Effective output up to 2048×2048; each dimension may range from 512–4096 within the total-pixel limit |
| Pixel Constraints | Total pixels must be between 768×768 and 2048×2048 |
| Aspect Ratio | 1:8 to 8:1 |
| Default Output Size | Typically matches the original image if not specified |
| Latency | Optimized for fast inference; no cold starts |
| Output Format | Image |
Sample Prompts
Change the jacket to black leather, keep the face and pose unchanged.Replace the background with a rainy Tokyo street at night, keep the subject unchanged, cinematic lighting.Change the product casing to brushed silver metal while preserving the original angle, lighting, and composition.
💰 Pricing
| Item | Price |
|---|---|
| Base price / per run | $0.03 |
💡 Best Use Cases
- Fashion and apparel: Swap clothing styles, colors, or materials while preserving model identity and pose.
- Background replacement: Change scene setting or mood without rebuilding the full image.
- Product image editing: Create color, material, and environment variants for e-commerce and marketing assets.
- Creative style iteration: Explore visual directions quickly while keeping the original composition intact.
🔗 Related Models
- Wan 2.7 Image Edit Pro: Higher-fidelity version with up to 2K editing output.
- Wan 2.6 Image Edit: Previous-generation image editing model with a similar prompt-driven workflow.
- Qwen Image Edit: General-purpose image editing model for everyday creative and product use cases.
- Google Nano Banana Pro Edit: High-fidelity editing model with strong composition preservation and text handling.
Related models
xAI Grok Imagine Image v2.0 Edit
xAI Grok Imagine Image V2.0 Edit transforms input images with text prompts, with configurable resolution and quality for precise image editing, visual refinements, creative variations, social content, marketing assets, and production workflows. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.
Qwen Image 3.0 Pro Edit | Fast Image Editing
Qwen Image 3.0 Pro Edit is a professional-grade image editing model with superior quality and advanced instruction understanding. Up to 2k. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.
Head Swap (Image)
Swap the head in a photo with the head from a reference image. Upload the target photo first, then the head reference.
Nami Qwen I2I Spicy
Edit a reference image with natural-language instructions while preserving the parts you do not ask to change.
Nano Banana Image Edit
Fast & Affordable Editing. Intuitive object replacement and style transfer without manual masking.
Nano Banana Pro Image Edit
4K Semantic Editing.Modifies existing images via natural language with context awareness and multilingual text editing.
Nano Banana 2 Image Edit
4K Semantic Editing.Modifies existing images via natural language with context awareness and multilingual text editing.
Seedream V5.0 Pro Edit API
Seedream V5.0 Pro Edit by ByteDance edits and generates images from single-image or multi-reference inputs, supporting up to 10 reference images, aspect ratio selection, and 1K / 2K output tiers. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.