Imagen

AI Video API for Developers

Text-to-video, image-to-video, video-to-video, lip sync and motion control on one REST endpoint. Kling, Veo, Wan, Seedance and Minimax behind one key. Video plans from $21/month for 3,250 API calls.

Get API Key — $21/Month API Documentation

What the AI Video API Covers

One Video API, Every Input Type

An AI video API usually means text-to-video. The ModelsLab video API is wider than that: a prompt, a still image, an existing clip or an audio track can each be the input, and the same key covers all of them. Text-to-video and image-to-video generate new footage; video-to-video restyles footage you have; lip sync matches a speaker’s mouth to a new audio track; motion control drives a subject with a reference motion.

The model is a parameter, not a separate integration. Open-weight models such as Wan and LTX run on ModelsLab GPUs and are covered by your plan. Third-party models — Kling, Veo, Seedance, Minimax, Runway, Vidu — sit behind the same endpoint and are billed per call from wallet balance, with the price shown on each model page before you send the request.

Built for Async

Video renders take seconds to minutes depending on model, duration and resolution. Short renders return an output URL in the response. Longer ones return an id: poll POST /api/v7/video-fusion/fetch/{id} until `output` is populated, or pass a `webhook` URL and ModelsLab posts the finished video to you.

text-to-video, image-to-video, video-to-video, lip-sync, motion-control
model_id selects Kling, Veo, Wan, Seedance, Minimax, LTX and more
Prompt, duration, aspect ratio and resolution controls
Async ids with a fetch endpoint and webhook delivery
One API key shared with image, speech and LLM
Video plans from $21/month for 3,250 API calls

Five Video Endpoints, One Key

Everything lives under /api/v7/video-fusion/. Pick the endpoint for the input you have; pick the model with model_id.

Endpoint	Input	Output
POST /api/v7/video-fusion/text-to-video	prompt, model_id, duration, aspect_ratio	Generated clip
POST /api/v7/video-fusion/image-to-video	init_image, prompt, model_id, duration	Animated clip from a still
POST /api/v7/video-fusion/video-to-video	init_video, prompt, model_id	Restyled clip
POST /api/v7/video-fusion/lip-sync	init_video, init_audio, model_id	Clip with matched mouth movement
POST /api/v7/video-fusion/motion-control	Reference motion + subject image	Subject driven by the motion
POST /api/v7/video-fusion/fetch/{id}	Request id	Status and output URL

Every endpoint returns an id for long renders; POST /api/v7/video-fusion/fetch/{id} to poll, or pass a webhook URL.

AI Video API quick start

Generate from a prompt, animate an image, poll for the result.

Text to video (Python)

Python

1import requests
2
3response = requests.post(
4    "https://modelslab.com/api/v7/video-fusion/text-to-video",
5    headers={
6        "Authorization": "Bearer YOUR_MODELSLAB_API_KEY",
7        "Content-Type": "application/json",
8    },
9    json={
10        # wan2.2 runs on ModelsLab GPUs — covered by your plan.
11        # Swap to "kling-v3-t2v" or "seedance-2-5-text-to-video" for a premium model.
12        "model_id": "wan2.2",
13        "prompt": "a lighthouse at dusk, waves breaking, slow push-in",
14        "duration": 5,
15        "aspect_ratio": "16:9",
16    },
17)
18
19print(response.json())

Image to video (JavaScript)

JavaScript

1const response = await fetch(
2  'https://modelslab.com/api/v7/video-fusion/image-to-video',
3  {
4    method: 'POST',
5    headers: {
6      'Content-Type': 'application/json',
7      Authorization: 'Bearer YOUR_MODELSLAB_API_KEY',
8    },
9    body: JSON.stringify({
10      model_id: 'wan2.2',
11      prompt: 'slow orbit around the product, soft studio light',
12      init_image: ['https://example.com/product-shot.jpg'],
13      duration: 5,
14    }),
15  },
16);
17
18const { id, output } = await response.json();
19console.log(output?.[0] ?? `queued as ${id}`);

Poll for a finished render (cURL)

bash

1# The fetch endpoint takes the request id as the last path segment.
2curl -X POST 'https://modelslab.com/api/v7/video-fusion/fetch/REQUEST_ID' \
3  -H 'Authorization: Bearer YOUR_MODELSLAB_API_KEY' \
4  -H 'Content-Type: application/json'
5
6# {"status":"success","output":["https://cdn.modelslab.com/.../clip.mp4"]}

What Video Plans Cost

Plans are billed per API call. Whether a call also draws wallet balance depends on the model.

Plan	Price / Month	API Calls	Third-party models
Basic	$21	3,250	Per call from wallet
Standard	$47	10,000	Per call from wallet
Unlimited Premium	$199	Unlimited on self-hosted models	Per call from wallet

ModelsLab video plan pricing, September 2026. Unlimited applies to the open-weight video endpoints ModelsLab self-hosts. Third-party video models (Kling, Veo, Seedance, Minimax, Runway, Vidu and others) are billed per call from wallet balance on any plan.

Ship Video in 3 Steps

From API key to a hosted MP4.

STEP 01

Step 1: Get Your API Key

Create a ModelsLab account, subscribe to a video plan from $21/month for 3,250 API calls, and copy your key from the dashboard.

STEP 02

Step 2: Pick the Endpoint and the Model

POST to text-to-video, image-to-video, video-to-video or lip-sync with a model_id. Open-weight models are covered by your plan; Kling, Veo and Seedance are billed per call from wallet balance.

STEP 03

Step 3: Fetch the Render

Short renders return an output URL. Longer ones return an id — poll /api/v7/video-fusion/fetch/{id} until output is populated, or pass a webhook URL.

Get API Key

Video API guides

Text to Video API

Endpoint reference for generating video from a prompt with Kling, Wan, Seedance and Veo.

Image to Video API

Animate a still image into a clip through one REST call.

Lip Sync API

Match mouth movement to a new audio track for dubbing and talking-head video.

Seedance API

Seedance 2.5, 2.0 and 1.5 Pro without a ByteDance Ark account.

Cheapest AI Video API

What video generation actually costs across plans and models.

Best AI Video API 2026

Veo, Kling and Seedance compared by quality, duration and price.

Why We Do Not Advertise “Unlimited Video”

The $199/month plan is unlimited on the open-weight video models ModelsLab self-hosts. It does not cover the third-party models, because Kling, Veo, Seedance and the rest bill ModelsLab per call, and we pass that through rather than hide it. If your pipeline runs on open-weight video, the unlimited plan removes per-call cost entirely; if it runs mostly on premium models, a metered plan plus wallet balance costs less.

Routing by Job

Draft, background and bulk renders belong on open-weight models covered by your plan. Premium third-party models earn their per-call price on the clips a customer actually sees. Both run on the same endpoint and key, so routing is a change to model_id and nothing else.

Why Developers Build Video on ModelsLab

Key advantages that set us apart

Text, image, video and audio inputs on one API

Kling, Veo, Wan, Seedance, Minimax, LTX by model_id

Async ids, fetch endpoint and webhook delivery

Open-weight models covered by your plan

Third-party prices shown per model, not hidden

No cold starts on self-hosted models

Video plans from $21/month for 3,250 calls

One API key across image, video, speech and LLM

Our Popular Use Cases

What teams ship with the AI video API:

Animate product stills into short loops for listings and ads with image-to-video, one request per SKU.

AI Video API FAQ

Your Data is Secure: GDPR Compliant AI Services

ModelsLab GDPR Compliance Certification Badge

GDPR Compliant

Start Creating Amazing AI Videos in Minutes

Choose plan as per your needs, cancel anytime.

Coming Soon

We are making some changes to our pricing, please check back later.

Get Expert Support in Seconds

We're Here to Help.

Want to know more? You can email us anytime at support@modelslab.com

View Docs

Text-to-video, image-to-video, video-to-video, lip sync and motion control, each as its own endpoint under /api/v7/video-fusion/. One API key and one plan cover all of them.

Open-weight models ModelsLab self-hosts such as Wan and LTX, and third-party models including Kling, Veo, Seedance, Minimax, Runway and Vidu. You pick the model with `model_id`; the endpoint stays the same.

Video plans start at $21/month for 3,250 API calls, with Standard at $47/month for 10,000 calls. Generating on the open-weight models ModelsLab self-hosts costs one call from your plan; third-party models are billed per call from wallet balance on top.

No. It is unlimited on the open-weight video models ModelsLab self-hosts. The third-party endpoints (Kling, Veo, Seedance, Minimax and others) are billed per call from wallet balance on every plan, because those providers bill per call.

Short renders return an output URL in the response. Longer renders return an id; POST /api/v7/video-fusion/fetch/{id} until `output` is populated, or pass a `webhook` URL and ModelsLab posts the result to you.

ModelsLab is a paid service. Video plans start at $21/month for 3,250 API calls. Creating an account is free, but generating video requires an active plan.

Yes. Image, video, speech and LLM generation share one API key. Each product category has its own plan, so a video plan covers video calls and an image plan covers image calls.