---
title: Wan 2.7 API — Text-to-Image, Image Edit & Video (2026)
description: Run Alibaba Cloud's Wan 2.7 models via REST API on ModelsLab: text-to-image, image edit, text-to-video, and image-to-video — all four checkpoints behind one key.
url: https://modelslab.com/wan-2-7
canonical: https://modelslab.com/wan-2-7
type: website
component: Seo/ModelPage
generated_at: 2026-06-04T13:45:03.182531Z
---

Available now on ModelsLab · Image Generation

Wan 2.7 — Text-to-Image, Edit, Video
All four Wan 2.7 checkpoints via one API key — no Alibaba Cloud account.
---

[Try Wan 2.7 Free](/models/alibaba_cloud/wan-2.7-t2i) [API Documentation](https://docs.modelslab.com)

Four checkpoints, one API surface
---

Text-to-Image

### High-quality image generation

Wan 2.7 T2I produces 1024×1024 images with strong prompt adherence — understands composition, lighting, and spatial relationships.

Image Edit

### Instruction-based editing

Swap backgrounds, change subjects, adjust style — all from a single text command via Wan 2.7 I2I.

Text-to-Video

### Generate video clips

Wan 2.7 T2V generates short clips with consistent motion and subjects. Preserves identity across frames.

Image-to-Video

### Animate still images

Wan 2.7 I2V turns a single image into a short clip with optional motion prompt for direction and speed.

Unified API

### One key, four checkpoints

All four Wan 2.7 variants share the same request/response shape. Swap model\_id to change capability.

Pricing

### 3-5x cheaper than Kling

~$0.10–$0.20 per 5s clip — typically 3–5x cheaper than Kling v1.5 or Seedance 2 for comparable workloads.

Examples

See what Wan 2.7 can do
---

Copy any prompt below and try it yourself in the [playground](/models/alibaba_cloud/wan-2.7-t2i).

Product photography

“a cinematic product shot of a red fox in snow, depth of field, golden hour light”

Background swap

“change the background to a neon-lit Tokyo street, keep the subject and pose identical”

Short clip from text

“a cat gracefully jumping across wooden rooftops at dusk, slow motion”

Animate a product photo

“rotate the product slowly 360 degrees, preserve lighting and background”

For Developers

A few lines of code.
Swap model\_id to change capability.
---

ModelsLab handles the infrastructure: fast inference, auto-scaling, and a developer-friendly API. No GPU management needed.

- **Serverless:** scales to zero, scales to millions
- **Pay per generation,** no minimums
- **Python and JavaScript SDKs,** plus REST API

[API Documentation ](https://docs.modelslab.com)

PythonJavaScriptcURL

Copy

```
<code>import requests

response = requests.post(
    "https://modelslab.com/api/v7/images/text-to-image",
    json={
  "key": "YOUR_API_KEY",
  "width": 1024,
  "height": 1024,
  "prompt": "a cinematic shot of a red fox in snow",
  "samples": 1,
  "model_id": "wan-2.7-t2i"
}
)
print(response.json())</code>
```

FAQ

Common questions about Wan 2.7 — Text-to-Image, Edit, Video
---

[Read the docs ](https://docs.modelslab.com)

### What is Wan 2.7?

Wan 2.7 is Alibaba Cloud's latest multimodal generation family. It ships four specialised checkpoints — text-to-image, image edit, text-to-video, and image-to-video — trained on a unified multimodal backbone with strong prompt adherence.

### How do I access Wan 2.7 via API?

Sign up for a free ModelsLab account, generate an API key, and POST to the relevant endpoint with the Wan 2.7 model_id (wan-2.7-t2i, wan-2.7-i2i, wan2.7-t2v, or wan2.7-i2v). No Alibaba Cloud account required.

### What does Wan 2.7 cost?

Wan 2.7 is priced around $0.10–$0.20 per 5-second video clip and similar per-image rates. This is typically 3–5× cheaper than Kling v1.5 or Seedance 2 for comparable workloads.

### Is Wan 2.7 free to try?

Yes. Every ModelsLab account starts with free credits — no credit card required. Use them to test all four Wan 2.7 variants before picking a paid plan.

### What is the difference between Wan 2.7 T2I and I2I?

Wan 2.7 T2I generates images from scratch using a text prompt. Wan 2.7 I2I (image edit) modifies an existing image using an instruction prompt (e.g. "change background to snowy mountains"). Both run through the same REST endpoint; only the model_id and request payload differ.

### How does Wan 2.7 compare to Kling and Seedance 2?

Wan 2.7 is the cheapest of the three and the only one with text-to-image + image-edit checkpoints in the same family. Kling v1.5 leads on cinematic motion quality; Seedance 2 offers native audio sync. All three are available on ModelsLab behind one key.

### Can Wan 2.7 generate long videos?

Each Wan 2.7 video call produces up to 5–10 seconds. For longer sequences, generate multiple clips and stitch them together with a transition model — a common pattern for 30-60s social videos.

### Does Wan 2.7 support webhooks?

Yes. Pass a webhook URL in the request body and ModelsLab POSTs the completed asset (image or video) to your endpoint when generation finishes — useful for async video jobs and high-throughput batch workflows.

Ready to create?
---

Start generating with Wan 2.7 — Text-to-Image, Edit, Video on ModelsLab.

[Try Wan 2.7 Free](/models/alibaba_cloud/wan-2.7-t2i) [API Documentation](https://docs.modelslab.com)

---

*This markdown version is optimized for AI agents and LLMs.*

**Links:**
- [Website](https://modelslab.com)
- [API Documentation](https://docs.modelslab.com)
- [Blog](https://modelslab.com/blog)

---
*Generated by ModelsLab - 2026-06-04*