Grok Imagine Video

New
Replicate
Video Generation

xAI video with native audio and lip-sync, up to 15s

Queue video with Grok Imagine Video
Video generation runs asynchronously — we'll queue a job and you can track it in your history.
Sign in to try this model with €5 free credits.
Sign in
Generates as an async job — typically 30 s to 2 min.
TL;DR·Last updated March 25, 2026

Grok Imagine Video is video generation AI model from Replicate, priced at €0.000 per 1M input tokens with a unknown context window.

Try Grok Imagine Video

Image References

Sign in to generate — 50 free credits on sign-up

Examples

See what Grok Imagine Video can generate

0:05

Fun

"Golden retriever catching frisbee in slow motion"

Pricing

Price per Generation
Per generation€1.50

API Integration

Use our OpenAI-compatible API to integrate Grok Imagine Video into your application.

Install
npm install railwail
JavaScript / TypeScript
import railwail from "railwail";

const rw = railwail("YOUR_API_KEY");

// Simple — just pass a string
const reply = await rw.run("grok-video", "Hello! What can you do?");
console.log(reply);

// With message history
const reply2 = await rw.run("grok-video", [
  { role: "system", content: "You are a helpful assistant." },
  { role: "user", content: "Explain quantum computing simply." },
]);
console.log(reply2);

// Full response with usage info
const res = await rw.chat("grok-video", [
  { role: "user", content: "Hello!" },
], { temperature: 0.7, max_tokens: 500 });
console.log(res.choices[0].message.content);
console.log(res.usage);
Specifications
Price
€1.50
Avg. latency
90.0s
Est. duration
2min
Developer
Replicate
Category
Video Generation
Supported Formats
mp4
Tags
audio
i2v
xai

Deep dive — xAI's Grok Imagine Video

About xAI
Founded 2023 · San Francisco Bay Area, USA

xAI was founded in March 2023 by Elon Musk and a team of researchers from OpenAI, DeepMind, Google, Microsoft and Tesla, including Igor Babuschkin, Tony Wu, Christian Szegedy and Jimmy Ba. The company's stated mission is to 'understand the true nature of the universe' and it operates closely with X (formerly Twitter) and Tesla. xAI launched Grok-1 in November 2023 (later open-sourced), followed by Grok-1.5, Grok-2 and Grok-3 with native image generation (Aurora) and image-to-video (Grok Imagine). xAI runs the Colossus supercluster in Memphis (originally 100k H100s, expanded to 200k+ by 2025) and was valued above $50B in late 2024.

Visit xAI →
Architecture
Latent video diffusion / DiT (closed)

Grok Imagine Video is xAI's image-to-video generation feature shipped inside the Grok app and on X (Twitter). It is a closed model and xAI has published almost no technical detail beyond it being a 'native video generation' system that complements the Aurora image generator. Industry reporting and behaviour suggest a latent video diffusion architecture with a transformer denoiser, conditioned on text embeddings (likely from a Grok-family text encoder) and on an initial image. The product targets short social-media clips (6-15 seconds) at HD resolution with optional NSFW 'spicy' mode that bypasses many of the moderation filters used by competitors. Audio support (sound effects, ambient noise) was added in late 2025. Training data are presumed to draw heavily from xAI's privileged access to X video uploads, though xAI have not confirmed this.

Parameters
Undisclosed
Context
unknown
What it can do
  • Image-to-video and text-to-video for short social clips
  • Tight integration with X (Twitter) and the Grok app
  • NSFW / 'spicy' mode (within posted guidelines) unlike most rivals
  • Native audio (sound effects, ambient) since late 2025
  • Fast generation tuned for Colossus hardware
  • Available to X Premium / SuperGrok subscribers
  • Strong meme and pop-culture coverage
  • Mobile-first UX
  • Best for: meme content, social shorts, viral clips, rapid prototyping.
Training & License

Undisclosed; presumed to include licensed footage, web video and X platform uploads. xAI have not released a model card or data sheet.

License: Proprietary commercial licence under X / xAI terms of service.

Known limitations
  • Short clip duration (typically 6-15 seconds)
  • No public model card or technical paper
  • Quality and consistency below Veo 3 / Sora 2 in independent tests
  • Subscription-gated, no general API at launch
  • Permissive moderation has drawn criticism

Frequently asked questions

Start using Grok Imagine Video today

Get started with free credits. No credit card required. Access Grok Imagine Video and 100+ other models through a single API.