Wan 2.5 API (Wan2.5 Preview) on APIMart runs Alibaba's Wan 2.5 video model with model ID wan2.5-preview. It turns a text prompt or a single image into a 5 or 10 second video at 480p, 720p or 1080p, and every clip comes with generated sound or your own audio track.
Service Highlights
Click to upload audio
WAV/MP3, 3-30s (max 15MB, duration ≤ 30s)
Video link valid for 72 hours
Transparent pricing with no hidden fees. Pay only for what you use.
| Variant | Tier | GoldCurrent price | Platinum | Diamond | Official | Your savings |
|---|---|---|---|---|---|---|
wan2.5-preview | 480P | 0.336Credits(~$0.0336)/sec | 0.3192Credits(~$0.0319)/sec | 0.3024Credits(~$0.0302)/sec | 0.42Credits(~$0.042)/sec | 20% |
| 720P | 0.664Credits(~$0.0664)/sec | 0.6308Credits(~$0.0631)/sec | 0.5976Credits(~$0.0598)/sec | 0.83Credits(~$0.083)/sec | 20% | |
| 1080P | 1.096Credits(~$0.1096)/sec | 1.0412Credits(~$0.1041)/sec | 0.9864Credits(~$0.0986)/sec | 1.37Credits(~$0.137)/sec | 20% |
* Actual costs are subject to final output.
Wan 2.5 API access on APIMart: turn a prompt or one image into a 5 or 10 second video at 480p, 720p or 1080p with sound in every clip, optional custom audio and per-second billing.
50K+
Active Users
99.9%
Uptime
2x
Faster
—
Cost Savings
Call the Wan 2.5 API on APIMart with one POST request. Generation is asynchronous: submit a task, then poll the task endpoint for the video URL.
Save your APIMart key in the APIMART_API_KEY environment variable and send it as Authorization: Bearer <key> on every request.
POST /v1/videos/generations with model wan2.5-preview, your prompt, duration, resolution and size. Add one image in image_urls for image to video. The task ID is in data[0].task_id.
Poll GET /v1/tasks/{task_id} until status is completed, then read the video URL from data.result.videos[0].url[0] and save the file to your own storage.
See the Wan 2.5 API reference for every parameter, size table and response field.
curl -X POST https://api.apimart.ai/v1/videos/generations \
-H "Authorization: Bearer $APIMART_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "wan2.5-preview",
"prompt": "A street musician plays violin under a bridge at night, warm lamplight, the music echoes softly, slow push-in",
"duration": 5,
"resolution": "720p",
"size": "16:9"
}'import os
import time
import requests
API_BASE = "https://api.apimart.ai"
headers = {"Authorization": f"Bearer {os.environ['APIMART_API_KEY']}"}
# 1. Submit the generation task
payload = {
"model": "wan2.5-preview",
"prompt": "A street musician plays violin under a bridge at night, warm lamplight, the music echoes softly, slow push-in",
"duration": 5,
"resolution": "720p",
"size": "16:9"
}
resp = requests.post(f"{API_BASE}/v1/videos/generations", headers=headers, json=payload, timeout=60)
resp.raise_for_status()
task_id = resp.json()["data"][0]["task_id"]
# 2. Poll the task until it finishes
for _ in range(200):
task_resp = requests.get(f"{API_BASE}/v1/tasks/{task_id}", headers=headers, timeout=30)
task_resp.raise_for_status()
task = task_resp.json()["data"]
if task["status"] == "completed":
print(task["result"]["videos"][0]["url"][0])
break
if task["status"] == "failed":
raise RuntimeError(task.get("error", {}).get("message", "generation failed"))
time.sleep(10)
else:
raise TimeoutError(f"task {task_id} did not finish in time; continue GET polling with this task ID")curl -X POST https://api.apimart.ai/v1/videos/generations \
-H "Authorization: Bearer $APIMART_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "wan2.5-preview",
"prompt": "The waves in the photo start rolling onto the shore, seagulls call in the distance",
"image_urls": [
"https://example.com/beach.jpg"
],
"duration": 5,
"resolution": "720p"
}'TASK_ID="task_xxx" # replace with the task_id returned by the POST request
curl "https://api.apimart.ai/v1/tasks/$TASK_ID" \
-H "Authorization: Bearer $APIMART_API_KEY"| Parameter | Type | Required | Default | Description |
|---|---|---|---|---|
model | string | Required | Default: wan2.5-preview | Use wan2.5-preview for Wan 2.5. |
prompt | string | Required | Default: — | Scene, motion and sound description. Optional but recommended for image to video. |
image_urls | array | Optional | Default: — | Exactly one image (public URL or Base64 data URI) for image to video. |
negative_prompt | string | Optional | Default: — | Content to avoid, up to 500 characters. |
resolution | string | Optional | Default: 720p | 480p, 720p or 1080p. |
size | string | Optional | Default: 16:9 | 16:9, 9:16, 1:1, 4:3 or 3:4 (480p supports the first three). Text to video only; sending it in image to video returns an error. |
duration | integer | Optional | Default: 5 | 5 or 10 seconds. |
audio | boolean | Optional | Default: true | Only true is supported; every video includes sound. |
audio_url | string | Optional | Default: — | wav or mp3, 3–30 seconds, up to 15 MB. Cut to the video length if longer. |
prompt_extend | boolean | Optional | Default: true | Rewrite and enrich the prompt. Helps short prompts but adds latency. |
seed | integer | Optional | Default: — | 0 or greater for more repeatable results. |
watermark | boolean | Optional | Default: false | Add an AI-generated mark in the bottom-right corner. |
nsfw_check | boolean | Optional | Default: false | Moderate the prompt and input image with omni-moderation-latest before submission. |
Compare every Alibaba Wan video model on APIMart. One API key calls all of them, billed per second of video.
| Model | Best for | Resolution | Duration |
|---|---|---|---|
| Wan 3.0 API | Longest Wan clips, with first and last frames or image, video, audio, document and webpage references | 480P / 720P / 1080P | 2–30 s |
| Wan 3.0 Prime API | Same parameters as Wan 3.0 under a separate model and price | 480P / 720P / 1080P | 2–30 s |
| Wan 2.7 API | Text to video, first and last frames, video continuation and audio-driven clips | 720P / 1080P | 2–15 s |
| Wan 2.7 R2V API | Keeping characters consistent from reference images or videos | 720P / 1080P | 2–15 s |
| Wan 2.7 Video Edit API | Editing or restyling an existing video with an instruction | 720P / 1080P | Follows source, up to 10 s |
| Wan 2.6 API | Multi-shot text to video and image to video with audio and effect templates | 720p / 1080p | 5 / 10 / 15 s |
| Wan 2.6 I2V API | Dedicated image to video from a single first frame | 720p / 1080p | 5 / 10 / 15 s |
| Wan 2.6 I2V Flash API | Image to video at any length from 2 to 15 seconds, with or without audio | 720p / 1080p | 2–15 s |
| Wan 2.5 APIThis page | Short clips that always include sound, with a low-cost 480p tier | 480p / 720p / 1080p | 5 / 10 s |
Guides, comparisons and pricing analysis from the APIMart blog.
What the wan2.5-preview model does in text-to-video and image-to-video mode
Workflows that use Wan 2.5 built-in audio and low-cost 480p drafts
From sign-up to your first Wan 2.5 video in three steps
Register on APIMart, add balance and create an API key in the dashboard.
Try prompts, images, durations and resolutions above and check the cost estimate before you write code.
Submit tasks from your backend with the example below, then poll for the result.
Real feedback from creators and developers using WAN 2.5 worldwide
“WAN 2.5 delivers great consistency! Character images remain stable across multiple clips, which was previously hard to achieve.”
Wei Zhang
Independent Animator
“As a developer, I value stability and speed. WAN 2.5 on APIMart delivers great performance with an easy-to-use API.”
David Chen
Full-Stack Engineer
“The image-to-video feature saved me so much post-production time. I just provide the keyframes and let WAN 2.5 do the rest.”
Ming Li
Illustrator
Answers about Wan 2.5 Preview, its audio, parameters, pricing and newer Wan models
Wan 2.5, released as Wan 2.5 Preview, is an Alibaba video model that turns text or a single image into a 5 or 10 second video at 480p, 720p or 1080p with generated sound. On APIMart its model ID is wan2.5-preview.
Alibaba previewed Wan 2.5 at its Apsara Conference in September 2025. The wan2.5-preview model is available through the APIMart Wan 2.5 API now.
It is one asynchronous endpoint, POST /v1/videos/generations with model wan2.5-preview. Leave out image_urls for text to video, or send one image for image to video; the mode is chosen automatically. You then poll the task for the video.
Yes. Every Wan 2.5 Preview video includes audio, and audio cannot be set to false. You can also supply your own wav or mp3 of 3–30 seconds with audio_url.
Wan 2.6 is newer and adds 15 second clips, multi-shot storytelling and effect templates. Wan 2.5 Preview is the only one of the two with a 480p tier and always includes sound, which makes it useful for low-cost drafts with audio.
Use Wan 2.5 Preview for 5 or 10 second clips with sound, including 480p drafts. Use Wan 2.6 for multi-shot clips up to 15 seconds, Wan 2.7 for first-last-frame control and video continuation, and Wan 3.0 for videos up to 30 seconds with multimodal references. The comparison table on this page links to each one.
model (wan2.5-preview) is required, and prompt is required for text to video. Optional fields are image_urls (one image), negative_prompt (up to 500 characters), resolution (480p, 720p or 1080p, default 720p), size (text to video only, default 16:9), duration (5 or 10, default 5), audio_url, prompt_extend (default true), seed, watermark and nsfw_check.
Yes. Send exactly one image in image_urls as a public URL or Base64 data URI and describe the motion in the prompt. Do not send size in image to video; the output follows the image's aspect ratio.
480p supports 16:9, 9:16 and 1:1, for example 832×480. 720p and 1080p also support 4:3 and 3:4; 16:9 is 1280×720 at 720p and 1920×1080 at 1080p.
Submit to POST /v1/videos/generations and read the task ID from data[0].task_id. Poll GET /v1/tasks/{task_id} every few seconds until status is completed, then read data.result.videos[0].url[0].
Wan 2.5 is billed per second of video, and the rate rises from 480p to 720p to 1080p. See the pricing table on this page for current rates; Gold members save 20%, and Platinum and Diamond members save up to 28%.
No to both. APIMart has no free quota for Wan 2.5; you pay per second for each task. Alibaba has not published Wan 2.5 weights, so there is no official download or app to run it locally. Alibaba's open-weight Wan releases are the Wan 2.2 family and specialized models such as Wan-Dancer; Wan 2.5 and later general video models are API-only.
Sign up on APIMart, add balance and create a key on the API Keys page. Send it as Authorization: Bearer <key>. The same key works for every Wan model and the rest of the APIMart catalog.
Yes. Wan2.5, Wan 2.5, Wan 2.5 Preview and Alibaba Wan 2.5 all refer to the same model, which APIMart exposes as wan2.5-preview.
Explore more models in the same category.

ViduQ 3 Turbo
ViduQ3-Turbo is the high-speed version of the Vidu Q3 series, featuring “faster and more affordable” AI video generation.

MiniMax Hailuo 2.3 Fast
MiniMax-Hailuo-2.3-Fast is an AI video generation model designed with a focus on speed.

Kling V3
Kling-V3 is an AI video model that balances quality and control, making it ideal for creating medium- to high-quality content.

Kling Video O1
Kling-Video-O1 is a multimodal AI video model that combines generation and editing, emphasizing high consistency and strong controllability.