Updated
What is the Grok Imagine Video API?
The Grok Imagine Video API is live on UnificAlly as xai/grok-imagine-1.5-fast-video, billed per second of video. The model key was renamed from xai/grok-imagine-video on 18 October 2026; the old key is rejected with a message naming the new one, and parameters and prices did not change.
Grok Imagine Video is a speed-optimized build of Grok Imagine 1.5 Video, the same model as the 1.5 Video API route, tuned for faster results in exchange for a small amount of quality. xAI's own figure is a 6-second 720p video in about 25 seconds.
It is also the more flexible build. Pass prompt alone for text-to-video, or supply either a single start_image_url or 1 to 7 image_urls for image-to-video, never both in the same request. Sound generates in the same pass, and audio: false turns it off. It is the only Grok Imagine route with style presets, with @ImageN referencing, and the only one whose videos can be extended or upscaled. On the Arena image-to-video board it still ranks ahead of every Veo 3.1 variant.
Grok Imagine Video API specs
| Model ID | xai/grok-imagine-1.5-fast-video |
| Endpoint | POST https://api.unifically.com/v1/tasks |
| Request types | Text-to-video, image-to-video from a start frame, reference images, saved characters, props, locations |
| Duration | 1 to 30 seconds, 6 by default |
| Resolution | 480p, 720p (default), 1080p |
| Aspect ratio | 1:1 (default), 2:3, 3:2, 9:16, 16:9 |
| Audio | On by default, audio: false for a silent clip |
| Reference inputs | One start_image_url, or 1 to 7 image_urls. Also character, prop and location references |
| Style presets | custom (default), spicy, fun, normal |
| Rate at 480p | $0.04/sec |
| Rate at 720p | $0.07/sec |
| Rate at 1080p | $0.125/sec |
Billing is per second of video, with no subscription. The pricing page lists the current rate for every model.
How to use the Grok Imagine Video API
- Create an API key on the API keys page.
- Send a POST request to
https://api.unifically.com/v1/taskswith the model ID and aninputobject. - Read the
task_idfrom the response, then check/v1/tasks/{task_id}until the task is finished, or set acallback_urlon the request and wait for the result.
curl -X POST https://api.unifically.com/v1/tasks \
-H "Content-Type: application/json" \
-H "Authorization: Bearer YOUR_API_KEY" \
-d '{
"model": "xai/grok-imagine-1.5-fast-video",
"input": {
"prompt": "A futuristic cityscape at sunset with flying cars",
"duration": 10,
"resolution": "1080p",
"aspect_ratio": "16:9",
"audio": false,
"video_preset": "custom"
}
}'
The response returns the task ID straight away:
{
"code": 200,
"success": true,
"data": {
"task_id": "abc123def456",
"status": "processing"
}
}
The finished task carries the video URL. That same task_id is what Extend and Upscale take.
curl https://api.unifically.com/v1/tasks/{task_id} \
-H "Authorization: Bearer YOUR_API_KEY"
Grok Imagine Video API parameters
| Parameter | Type | Default | What it does |
|---|---|---|---|
prompt | string | none | Video prompt, required. Use @ImageN for an item in image_urls, or @audio1 for the request-level voice |
aspect_ratio | string | 1:1 | Output aspect ratio: 1:1, 2:3, 3:2, 9:16, or 16:9 |
duration | integer | 6 | Video duration in seconds, 1 to 30 |
resolution | string | 720p | Output resolution: 480p, 720p, or 1080p |
start_image_url | string | none | Start frame image URL. Cannot be combined with image_urls or media references |
image_urls | string[] | none | 1 to 7 ordered reference images. Cannot be combined with start_image_url |
character_id | string | none | ID of an existing saved Grok character reference |
character | object | none | Transient character reference: 3 images, or 2 images plus a preset voice |
prop | object | none | Transient prop reference containing a name and 1 to 3 images |
location | object | none | Transient location reference containing a name and 1 to 3 images |
voice | string | none | Case-insensitive preset voice name. Audio URLs are not accepted. Needs a visual input and cannot be combined with audio: false |
audio | boolean | true | Generates audio. false renders a silent clip |
video_preset | string | custom | Generation style preset: custom, spicy, fun, or normal |
nsfw_check | boolean | false | Checks the prompt and input images for NSFW content before generating. Off unless set to true |
The Grok Imagine Video API documentation has the full reference, with the rules for character, prop and location objects and more request examples.
Image to video with generated audio
Send start_image_url for a plain image-to-video request, never alongside image_urls:
curl -X POST https://api.unifically.com/v1/tasks \
-H "Content-Type: application/json" \
-H "Authorization: Bearer YOUR_API_KEY" \
-d '{
"model": "xai/grok-imagine-1.5-fast-video",
"input": {
"prompt": "Camera slowly zooms in while clouds move across the sky",
"start_image_url": "https://example.com/landscape.jpg",
"audio": true,
"duration": 6,
"resolution": "720p",
"aspect_ratio": "16:9"
}
}'
Reference images
@Image1 and @Image2 point at the uploads by position:
curl -X POST https://api.unifically.com/v1/tasks \
-H "Content-Type: application/json" \
-H "Authorization: Bearer YOUR_API_KEY" \
-d '{
"model": "xai/grok-imagine-1.5-fast-video",
"input": {
"prompt": "The woman in @Image1 walks into the neon-lit bar from @Image2",
"image_urls": [
"https://example.com/woman.jpg",
"https://example.com/neon-bar.jpg"
],
"audio": false,
"duration": 10,
"resolution": "720p",
"aspect_ratio": "16:9"
}
}'
Key features of Grok Imagine Video
Referencing assets in the prompt
Address each upload by position: "@Image1 does a dance in @Image2". Numbering follows the order of image_urls, so @Image1 is the first image in the array. Only this build understands the syntax.
Fast turnaround
A 6-second 720p video comes back in about 25 seconds by xAI's figure, down from 40 or more on the previous model. Built for iterating on direction, not waiting on it.
Ahead of Veo 3.1 on the Arena
Even as the speed build, it ranks above every Veo 3.1 variant on the Arena image-to-video board as of June 2026. Boards move, so read it as a current snapshot.
NSFW and style presets
Spicy, Fun and Normal presets and NSFW output are only on this build. The official 1.5 Video API build is safe-for-work with no presets.
Up to 7 reference images
Lock product, talent, and background separately while the prompt drives camera and action. Or pass a single start_image_url when you only need an opening frame.
The only build that extends and upscales
Grok Imagine Video Extend and Grok Imagine Upscale accept task IDs from this route only, which is how a Grok sequence runs past 30 seconds or reaches HD.
What Grok Imagine 1.5 Fast Video is best for
Text-to-video from a brief
Generate a video from a prompt alone when there is no source image to work from.
Choreographed two-subject shots
Put a subject into a scene explicitly: "@Image1 walks through @Image2" beats describing both from scratch.
Multi-reference brand spots
Separate product, talent, and background across up to 7 references, then direct them by name.
Fast iteration on direction
Explore camera, pacing, and composition with quick turnarounds before committing.
Sequences past 30 seconds
Generate an opening here, then chain Extend in 6 or 10-second steps.
Preset-driven styling
Reach for Spicy, Fun, or Normal when the look matters more than a precisely worded prompt.
Grok Imagine 1.5 Fast Video use cases
Referencing assets is what sets this build apart. Upload a dancer and a stage, then prompt "@Image1 does a slow dance in @Image2", and the model knows which upload is the subject and which is the scene instead of guessing from a description. A brand team can lock a product, a presenter, and a backdrop across three references and direct them by name in one call, with a matching sound bed generated alongside.
Because Extend and Upscale only accept task IDs from this route, it is also the entry point for anything longer or sharper than one pass: generate here, chain Extend for length, then Upscale the result you keep. The alternative path is to settle the opening frame here and re-animate it on 1.5 Video at 1080p.
Grok Imagine 1.5 Fast Video limitations
Output tops out at 1080p. Being speed-optimized, it trades a small amount of quality against that build at the same prompt. A single pass caps at 30 seconds, so going longer means chaining Extend. Negative prompts are ignored, so describe what you want rather than what to avoid. A start image and reference images are mutually exclusive: send one field or the other, never both, and image_urls accepts a maximum of 7.
Grok Imagine Video vs Grok Imagine 1.5 Video
| Grok Imagine Video | Grok Imagine 1.5 Video | |
|---|---|---|
| Build | Speed-optimized | Official API |
| Text-to-video | Yes | Yes |
| Duration | 1–30s | 1–15s |
| Resolution | 480p, 720p, 1080p | 480p, 720p, native 1080p |
| Images | One start_image_url or 1–7 image_urls, plus characters, props, locations | One start_image_url or 1–14 image_urls, plus end frame and keyframes |
| Referencing assets | @Image1, @Image2 in the prompt | None |
| NSFW | Via presets | No |
| Style presets | Custom, Spicy, Fun, Normal | None |
| Extend / Upscale | Supported | Not supported |
| Billing key | xai/grok-imagine-1.5-fast-video | xai/grok-imagine-1.5-video |
Pick this build for NSFW, presets, characters, 30 second clips, or extend and upscale by task ID. Pick 1.5 Video for the API's control set: 14 references, end frame, keyframes, voices and native 1080p, safe-for-work.
FAQs
People also ask about Grok Imagine 1.5 Fast Video
It is a speed-optimized build of xAI's Grok Imagine 1.5 Video model, faster results in exchange for a small amount of quality. It is the only Grok Imagine route with style presets, saved characters, props, locations, and @ImageN referencing.
Fast video tasks for xai/grok-imagine-1.5-fast-video need a key from the API keys page, sent as Authorization: Bearer <key> to https://api.unifically.com/v1/tasks. The task ID you get back is also what Extend and Upscale take, under the same key.
Pass a prompt alone for text-to-video, or add a start_image_url or 1 to 7 image_urls (never both), with duration up to 30 seconds, resolution, aspect_ratio, audio and video_preset. The request is queued as a task; the finished task carries the video URL, read from /v1/tasks/{task_id} or sent to a callback_url.
The Grok Imagine Video API is billed per second of video, with no subscription. The rate is $0.04/sec at 480p, $0.07/sec at 720p, and $0.125/sec at 1080p. The rates shown here follow the pricing page automatically.
It was renamed to xai/grok-imagine-1.5-fast-video on 18 October 2026 so the name says which version you get. Requests to the old key are rejected with a message pointing at the new one, and nothing is charged. Parameters and prices are unchanged, and Extend and Upscale accept task IDs from the new key.
xAI's own figure for the Fast build is a 6-second 720p video in about 25 seconds, down from 40 or more on the previous model. Actual times vary with duration, resolution, and load.
Same 1.5 model, two builds. This one is the grok.com build: NSFW and style presets, saved characters, props and locations, clips up to 30 seconds, and the only build Extend by task ID and Upscale accept. Grok Imagine 1.5 Video is the official API build: safe-for-work, native 1080p, up to 14 references, end frame, keyframes and preset voices, capped at 15 seconds.
Yes. Sound effects, ambience, and lip-synced dialogue generate in the same pass as the video. Audio is on by default, and setting audio to false renders a silent clip. Direct the sound through the prompt alongside the motion.
Yes. Reference your uploads by position with @Image1, @Image2, and so on, for example "@Image1 does a dance in @Image2". The numbering follows the order of image_urls, so @Image1 is the first image in the array.
Either a single start_image_url or 1 to 7 image_urls, one or the other, never both in the same request. Both are optional; omit them for text-to-video.
Custom, which is the default, plus Spicy, Fun, and Normal. Presets are only on this build; Grok Imagine 1.5 Video has none.
From 1 to 30 seconds, with a default of 6. This is the longest single pass of any Grok Imagine route. To go longer still, chain Grok Imagine Video Extend onto the result in 6 or 10-second steps.
Yes, and this is the only Grok Imagine build that can. Both Grok Imagine Video Extend and Grok Imagine Upscale accept task IDs from this route, and only from this route.
Grok Imagine 1.5 Fast Video guides and comparisons
Review pricing, limits, and tested outputs before running this model.
