Models
xAI
Grok Imagine Video

xAI

Grok Imagine Video

AI Video Generation Model

Mode:

All Text to Video Image to Video Video to Video

Video #5 Image-to-Image

xAI's video generation model based on the Aurora architecture, supporting text-to-video, image-to-video, and video editing with native audio-visual synthesis at up to 720p

Try Model

15 seconds

Max Video Duration

Supported Modes

Text to Video Image to Video Video to Video

Active

Details

Model ID

grok-imagine-video

Also known as: grok-video

Creator

xAI

Family

grok-imagine

Released

February 2026

Ready to integrate?

Access grok-imagine-video via our unified API.

Create Account

Available at 1 provider

Starting from

$0.050 /second via Replicate

Prices shown are in USD

Full pricing details

Provider Performance

Fastest generation through replicate at 379ms median latency with 100.0% success rate.

Aggregated from real API requests over the last 30 days.

Generation Time

replicate

p95: 1,759ms

Success Rate

replicate

100.0%

101 / 101 requests

Provider Rankings

#	Provider	p50 Gen Time	p95 Gen Time	Success Rate	TTFB (p50)
1	replicate	379ms	1,759ms	100.0%	—

Data updated every 15 minutes. Based on all API requests through Lumenfall over the last 30 days.

Providers & Pricing (1)

Grok Imagine Video is available exclusively through Replicate, starting at $0.05/video.

Replicate

Text to Video Image to Video Video to Video

replicate/grok-imagine-video

Provider Model ID: xai/grok-imagine-video

$0.050 /second

View pricing details

grok-video API Async video generation

Lumenfall provides an OpenAI-compatible API for generating 720p videos, performing image-to-video transformations, and executing video edits using Grok Imagine Video.

Base URL

https://api.lumenfall.ai/v1

Model

grok-imagine-video

Code Examples

Image to Video

/v1/videos/generations

# Step 1: Submit image-to-video request

VIDEO_ID=$(curl -s -X POST \

  https://api.lumenfall.ai/v1/videos \

  -H "Authorization: Bearer $LUMENFALL_API_KEY" \

  -H "Content-Type: application/json" \

  -d '{

    "model": "grok-imagine-video",

    "prompt": "",

    "image_url": "https://example.com/start-frame.jpg",

    "duration": "10",

    "aspect_ratio": "16:9"

  }' | jq -r '.id')

echo "Video ID: $VIDEO_ID"

# Step 2: Poll for completion

while true; do

  RESULT=$(curl -s \

    https://api.lumenfall.ai/v1/videos/$VIDEO_ID \

    -H "Authorization: Bearer $LUMENFALL_API_KEY")

  STATUS=$(echo $RESULT | jq -r '.status')

  echo "Status: $STATUS"

  if [ "$STATUS" = "completed" ]; then

    echo $RESULT | jq -r '.output.url'

    break

  elif [ "$STATUS" = "failed" ]; then

    echo $RESULT | jq -r '.error.message'

    break

fi

  sleep 5

done

const BASE_URL = 'https://api.lumenfall.ai/v1';

const API_KEY = 'YOUR_API_KEY';

// Step 1: Submit image-to-video request

const submitRes = await fetch(`${BASE_URL}/videos`, {

  method: 'POST',

  headers: {

    'Authorization': `Bearer ${API_KEY}`,

    'Content-Type': 'application/json'

},

  body: JSON.stringify({

    model: 'grok-imagine-video',

    prompt: '',

    image_url: 'https://example.com/start-frame.jpg',

    duration: '10',

    aspect_ratio: '16:9'

})

});

const { id: videoId } = await submitRes.json();

console.log('Video ID:', videoId);

// Step 2: Poll for completion

while (true) {

  const pollRes = await fetch(`${BASE_URL}/videos/${videoId}`, {

    headers: { 'Authorization': `Bearer ${API_KEY}` }

});

  const result = await pollRes.json();

  if (result.status === 'completed') {

    console.log('Video URL:', result.output.url);

    break;

  } else if (result.status === 'failed') {

    console.error('Error:', result.error.message);

    break;

  await new Promise(r => setTimeout(r, 5000));

import requests

import time

BASE_URL = "https://api.lumenfall.ai/v1"

API_KEY = "YOUR_API_KEY"

HEADERS = {

    "Authorization": f"Bearer {API_KEY}",

    "Content-Type": "application/json"

# Step 1: Submit image-to-video request

response = requests.post(

    f"{BASE_URL}/videos",

    headers=HEADERS,

    json={

        "model": "grok-imagine-video",

        "prompt": "",

        "image_url": "https://example.com/start-frame.jpg",

        "duration": "10",

        "aspect_ratio": "16:9"

video_id = response.json()["id"]

print(f"Video ID: {video_id}")

# Step 2: Poll for completion

while True:

    result = requests.get(

        f"{BASE_URL}/videos/{video_id}",

        headers=HEADERS

    ).json()

    if result["status"] == "completed":

        print(f"Video URL: {result['output']['url']}")

        break

    elif result["status"] == "failed":

        print(f"Error: {result['error']['message']}")

        break

    time.sleep(5)

Parameter Reference

Required Supported Not available

Core Parameters

Parameter	Type	Description	Modes
`prompt`	string	Required. Text prompt for video generation	T2V I2V V2V
`duration`	number	Video duration in seconds	T2V I2V V2V

Size & Layout

Parameter	Type	Description	Modes
`size`	string	Video dimensions as WxH pixels (e.g. "1920x1080") or aspect ratio (e.g. "16:9") `auto` `1365x768` `768x1365` `1254x836` `836x1254` `887x1182` `1024x1024` `1183x887` WxH determines both shape and scale (aspect_ratio and resolution are ignored when size is provided). W:H format is equivalent to aspect_ratio.	T2V I2V V2V
`aspect_ratio`	string	Aspect ratio of the output video (e.g. "16:9", "1:1") `auto` `9:16` `2:3` `3:4` `1:1` `4:3` `3:2` `16:9` Controls shape independently of scale. Use with resolution to control both. If size is also provided, size takes precedence. Any ratio is accepted and mapped to the nearest supported value.	T2V I2V V2V
`resolution`	string	Output resolution tier (e.g. "1K", "4K") `auto` `1K` Controls scale independently of shape. Higher tiers produce larger videos and cost more. If size is also provided, size takes precedence for scale. Any tier is accepted and mapped to the nearest supported value.	T2V I2V V2V

Output	`size`		`aspect_ratio` + `resolution`
Flexible
Auto	`"auto"`	—	Model chooses optimal dimensions

1K 7 sizes

Output	`size`		`aspect_ratio` + `resolution`
1183 × 887	`"1183x887"`	or	`"4:3"` + `"1K"`
1024 × 1024	`"1024x1024"`	or	`"1:1"` + `"1K"`
887 × 1182	`"887x1182"`	or	`"3:4"` + `"1K"`
836 × 1254	`"836x1254"`	or	`"2:3"` + `"1K"`
1254 × 836	`"1254x836"`	or	`"3:2"` + `"1K"`
768 × 1365	`"768x1365"`	or	`"9:16"` + `"1K"`
1365 × 768	`"1365x768"`	or	`"16:9"` + `"1K"`

How these parameters work

size

Exact pixel dimensions

"1920x1080"

aspect_ratio

Shape only, default scale

"16:9"

resolution

Scale tier, preserves shape

"1K"

Priority when combined

size › aspect_ratio + resolution › aspect_ratio › resolution

size is most specific and always wins. aspect_ratio and resolution control shape and scale independently.

How matching works

Shape matching – we pick the closest supported ratio. Ask for 7:1 on a model with 4:1 and 8:1, you get 8:1.

Scale matching – providers use different tier formats: K tiers (0.5K 1K 2K 4K) or megapixel tiers (0.25 1). If the exact tier isn't available, you get the nearest one.

Dimension clamping – if a model has pixel limits, we clamp dimensions to fit and keep the aspect ratio intact.

Media Inputs

Parameter	Type	Description	Modes
`input_reference`	array	Required for I2V. Input image(s) to animate into video	T2V I2V V2V

Output & Format

Parameter	Type	Description	Modes
`n`	integer	Number of videos to generate Default: `1` Gateway generates multiple videos in parallel even if provider only supports 1.	T2V I2V V2V

Parameter Normalization

How we handle parameters across different providers

Not every provider speaks the same language. When you send a parameter, we handle it in one of four ways depending on what the model supports:

Behavior	What happens	Example
`passthrough`	Sent as-is to the provider	style, quality
`renamed`	Same value, mapped to the field name the provider expects	prompt
`converted`	Transformed to the provider's native format	size
`emulated`	Works even if the provider has no concept of it	n, response_format

Parameters we don't recognize pass straight through to the upstream API, so provider-specific options still work.

Full API Reference Authentication, endpoints, and more

Grok Imagine Video FAQ

How much does Grok Imagine Video cost?

Grok Imagine Video starts at $0.05 per video through Lumenfall. Pricing varies by provider. Lumenfall does not add any markup to provider pricing.

How do I use Grok Imagine Video via API?

You can use Grok Imagine Video through Lumenfall's OpenAI-compatible API. Send requests to the unified endpoint with model ID "grok-imagine-video". Code examples are available in Python, JavaScript, and cURL.

Which providers offer Grok Imagine Video?

Grok Imagine Video is available through Replicate on Lumenfall. Lumenfall automatically routes requests to the best available provider.

Overview

Grok Imagine Video is a high-fidelity video generation model developed by xAI, designed to produce video content with integrated, synchronized audio. Built on the proprietary Aurora architecture, the model supports multiple input modalities including text-to-video, image-to-video, and direct video editing. It is distinctive for its native audio-visual synthesis, meaning the model generates temporal audio that aligns with the visual motion rather than adding a separate soundtrack post-generation.

Strengths

Native Audio-Visual Synthesis: The model generates synchronized spatial and temporal audio concurrently with the video frames, ensuring that sound effects match the actions on screen.
Multi-Modal Flexibility: It handles three distinct workflows (text prompts, image-to-video animation, and video-to-video editing) within a single unified framework.
Temporal Consistency: The Aurora architecture maintains stable object identity and physical logic across sequences, reducing the “morphing” artifacts common in earlier generation diffusion models.
Resolution and Quality: Capable of outputting video at up to 720p resolution, balancing computational efficiency with visual detail suitable for social media and web content.

Limitations

Resolution Ceiling: While 720p is effective for many applications, it falls short of the 1080p or 4K standards required for cinematic production or high-end commercial use.
Action Duration: Like most current generative video models, it is optimized for short-form clips; maintaining narrative or structural coherence over extended durations (e.g., several minutes) remains a challenge.
Inference Latency: The computational demands of native audio-visual generation may result in longer wait times compared to visual-only models.

Technical Background

Grok Imagine Video is built on xAI’s Aurora architecture, a specialized framework designed for high-dimensional temporal data. Unlike models that layer audio on top of finished video, Aurora utilizes a joint embedding space that treats audio and video as co-dependent signals during the generation process. This approach allows the model to “understand” the relationship between physics, motion, and sound within a single transformer-based or diffusion-variant bottleneck.

Best For

This model is ideal for creating short-form social media content, rapid prototyping for advertisements, and animating still photography with realistic environmental sound. It is a strong choice for developers who need “all-in-one” asset generation where visual motion and audio cues must be perfectly aligned without manual editing. Grok Imagine Video is available for testing and integration through Lumenfall’s unified API and playground, allowing for seamless incorporation into automated media pipelines.

Try Grok Imagine Video in Playground

Generate images with custom prompts — no API key needed.

Open Playground