Grok Imagine Video 1.5Text, Image, and Reference to Video, With Sound
Generate cinematic video with native audio using xAI's Grok Imagine Video 1.5 on fal. Start from a prompt, a still image, or a set of references, at up to 1080p and 15 seconds, with synchronized sound, lifelike physics, and strong subject consistency through a fast serverless API.
What Makes Grok Imagine Video 1.5 Different
Video and Sound in a Single Pass
Grok Imagine Video 1.5 generates synchronized audio natively alongside the video: sound effects timed to the on-screen action, ambient sound, music, and dialogue with accurate lip-sync. The whoosh of a blade, footsteps, and room tone all land on cue, so a single prompt returns production-ready video with cinema-grade sound in one pass, no separate post-production audio tools required.
Lifelike Motion, Physics, and Detail
Grok Imagine Video 1.5 expands a single image into a full scene with coherent motion, realistic physics, and fine detail: fluid dynamics, rising steam, and translucent materials like glass. It preserves the look of your source frame and follows prompts closely, with dynamic camera control through natural language, at up to 1080p.
Prompt It, Animate It, or Match It
Video 1.5 runs three modes on the same model. Text-to-video builds a clip from a prompt alone at up to 1080p. Image-to-video animates a still you supply as the first frame, also up to 1080p. Reference-to-video takes 1 to 7 reference images, tagged in the prompt as <IMAGE_0> onward, to carry a face or product through the shot without locking the opening frame. Pick the entry point that matches what you already have.
Start building with the Grok Imagine Video 1.5 API
Three modes on one model: generate from a prompt, animate a still, or match a set of references. All with native audio, all via API.
Generate videos from images with audio using xAI's Grok Imagine 1.5 Video model.
Generate videos from prompts with audio using xAI's Grok Imagine 1.5 Video model.
Generate videos from images and audio references using xAI's Grok Imagine 1.5 Video model.
See what Grok Imagine Video 1.5 can create
Turn on audio to hear the native sound generation. Each clip below was generated in one pass with no post-production.
UGC product spot with timed beats and synced dialogue
"(0-5s) Medium shot, she speaks warmly while gesturing to a product. (5-10s) Slow push-in to a close-up, glowing skin and expressive eyes. (10-15s) Cut to an over-shoulder framing of the vanity as she smiles. Glossy, warm, cinematic."
Product hero shot with a slow spin
"Brightly colored athletic running shoe resting on mossy ground, red to yellow gradient upper with grid pattern, thick sculpted neon green foam sole, red laces, wavy yellow eyestay overlay, surrounded by green moss and small ferns, blurred tree branches and bright blue sky background, extreme low angle close-up, vibrant product photography, sharp natural sunlight and doing a slow spin"
Landscape still animated with an orchestral score
"Camera tracks forward over the fjord as mist begins to drift between the mountains and the water ripples. A small red boat starts moving across the frame. Orchestral strings swell as the shot rises"
Cinematic scene with motion and ambient audio
"The character turns toward the camera and looks up, and behind him rain starts to fall. Crisp rainfall and fishing town ambience"
How to access the Grok Imagine Video 1.5 API
The client API handles the request submit protocol. It will handle the request status updates and return the result when the request is completed.
import { fal } from "@fal-ai/client";
const result = await fal.subscribe("xai/grok-imagine-video/v1.5/image-to-video", {
input: {
image_url: "https://example.com/your-image.jpg",
prompt: "Slow push-in as she smiles and speaks warmly, soft ambient room tone",
resolution: "720p",
duration: 10,
},
logs: true,
onQueueUpdate: (update) => {
if (update.status === "IN_PROGRESS") {
update.logs.map((log) => log.message).forEach(console.log);
}
},
});
console.log(result.data);
console.log(result.requestId);
Common questions about Grok Imagine Video 1.5
What is Grok Imagine Video 1.5?
Grok Imagine Video 1.5 is xAI's current video generation model. It runs three modes: text-to-video from a prompt alone, image-to-video from a still you supply as the first frame, and reference-to-video from a set of reference images. Every mode returns native audio in the same pass, covering dialogue, ambient sound, and effects, all synchronized without post-production.
What resolutions and durations does Grok Imagine Video 1.5 support?
Text-to-video and image-to-video run at 480p, 720p, or 1080p. Reference-to-video runs at 480p or 720p. Every mode accepts a duration from 1 to 15 seconds, with audio generated natively alongside the video. Text-to-video and reference-to-video accept seven aspect ratios, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, and 1:1, while image-to-video follows the aspect ratio of your source image.
How does image-to-video work?
You provide a starting image and a text prompt describing the motion, camera direction, and audio you want. The model expands the image into a full scene with coherent motion, realistic physics, and fine detail such as micro-expressions, eye tracking, and translucent materials, while preserving the subject's identity and staying visually consistent with your source frame. This is useful for animating still concepts, product shots, or reference frames into full video sequences with sound.
What makes Grok Imagine Video 1.5 stand out?
It pairs strong video quality with native audio generation, so a single pass produces video and synchronized sound together. Reference-to-video adds up to seven reference images, so a face, product, or outfit can be carried across a whole series of clips. And because all three modes share one model and one API key, you can move between prompt-first, image-first, and reference-first workflows without switching endpoint families.
How good is the audio?
Audio is generated natively alongside the video, so it stays in sync without post-production. Grok Imagine Video 1.5 produces natural dialogue with accurate lip-sync, contextually appropriate ambient sound, and well-timed effects, giving each clip a finished, cinematic feel.
What is reference-to-video?
Reference-to-video takes 1 to 7 reference images and carries the people, products, and wardrobe in them through the generated clip. Unlike image-to-video, the references guide what appears without locking the first frame, so you keep the subject and still direct the opening shot. Tag each image in your prompt as <IMAGE_0> through <IMAGE_N> so the model knows what each one is for. References are images only; there is no reference audio input, though the clip still comes back with natively generated audio. It is the most direct way to keep one character consistent across a series of clips.
How much does Grok Imagine Video 1.5 cost on fal.ai?
Pricing is pay-per-use with no minimums or subscriptions. Video is priced per second of output: $0.08/s at 480p and $0.14/s at 720p across all three modes, and $0.25/s at 1080p on text-to-video and image-to-video. Input images add $0.01 each, and reference-to-video accepts 1 to 7 of them. The audio the model generates is included at no extra charge. A 10-second 720p clip from a single image costs about $1.41.
How do I get started with the API?
Install the fal.ai SDK (Python or JavaScript), grab an API key from your dashboard, and make your first request in a few lines of code. The API is serverless, so there are no GPUs to manage and no infrastructure to set up. Check the API documentation for all available parameters.
Can I use Grok Imagine Video 1.5 for commercial projects?
Yes. Content generated through the fal.ai API can be used in commercial projects. Check fal.ai's terms of service for full details on usage rights and licensing.
Get in touch about Grok Imagine Video 1.5
Want to learn more about integrating Grok Imagine Video 1.5 into your workflow? Leave your details and our team will reach out.