KovAPIKovAPI
Video ● Maintenance Service status →

Grok Imagine Video 1.5

⧉
grok-imagine-video-1.5

xAI Grok Imagine Video 1.5: 4-15 seconds in one-second steps, 480P / 720P, five aspect ratios. Text-to-video without a reference image, image-to-video with reference images (up to 7; with 2 or more the maximum is 10 seconds). Asynchronous, billed per clip -- 1 second costs the same as 15 -- and failed jobs are refunded in full. 480P $0.30 / 720P $0.40; the price only changes with resolution.

Text-to-videoImage-to-videoPer-clip pricingAsync jobs

Live Test · Playground

Try out Grok Imagine Video 1.5 right here (available after login).

Input

0 / 5000 bytes
Reference images0 / 7
Click to upload or drag files here jpg / png / webp / gif, up to 16 MB each or paste links
With multiple references, address them in the prompt as @Image1 / @Image2, e.g. "the character from @Image1 stands in the scene from @Image2". Without tokens the model decides on its own.
Uploaded references are kept for 7 days.
Rendering takes ~1-3 min; billed per clip (any length, same price), fully refunded on failure. Prompt limit 5000 bytes (about 1666 Chinese characters).

Output

video
Results will appear here after running.
🕑 Generated images and videos are kept for 15 days before 2026-10-29 09:49:40 UTC; from that time, they are kept for 7 days and existing files older than 7 days are removed.

About Grok Imagine Video 1.5

Grok Imagine Video 1.5 is xAI's video model and the cheapest video model on KovAPI. What sets it apart from everything else here is the billing shape: it is priced PER CLIP, so a 1-second and a 15-second render cost exactly the same; only the resolution changes the price (480P $0.30, 720P $0.40). Duration is any integer 1-15 seconds, resolution 480P / 720P, aspect ratio 16:9 / 9:16 / 1:1 / 4:3 / 3:4; omit the reference image for text-to-video, include up to 7 for image-to-video; every clip ships with a native audio track. Async job: POST /v1/videos/generations returns HTTP 202 and a job id immediately; poll GET /v1/videos/jobs/{id} until status=succeeded and read data[0].url. That link is an mp4 re-hosted by us -- it is kept for 15 days (far longer than the upstream CDN link), so download it or copy it to your own storage before then.

Use cases

Batch idea validation

The lowest fixed cost per clip on the platform, so you can run dozens of prompts and keep the best one.

Long clips are the bargain

Per-clip pricing means the 15-second option is the best value; there is no reason to cut a clip short to save money.

Vertical social assets

9:16 straight out of the model for short-video platforms; 4:3 and 3:4 suit mixed image-and-text layouts.

Multi-image references

Up to 7 reference images, useful for putting one character or product into different scenes.

How to choose

Pick it when budget comes first or when you are batching ideas -- it starts at $0.30, the cheapest video here, and per-clip pricing means 15 seconds costs no more than one. For a controllable soundtrack, reference audio or reference video use the Seedance family; for more than 15 seconds use seedance-2.5; for 1080P use minimax-h3.

FAQ

Why per clip instead of per second?
Because the upstream charges per call: the cost is identical whatever length you ask for. We match the shape of our sell price to the shape of our cost rather than copying a per-second number that would overcharge short clips and lose money on long ones. Practically: within your chosen resolution, just ask for the longest duration you can use. The two resolutions cost us the same upstream; the split exists so that people who want the cheaper option have one.
Why only 10 seconds with multiple images?
That is an upstream limit: 15 seconds with a single reference image, 10 seconds with 2-7. We reject it at submit time with a clear message rather than letting you wait several minutes for a failure.
Does it produce audio?
Yes -- every measured render carries a native AAC stereo track. But the upstream endpoint has no switch and accepts no reference audio, so you can neither turn it off nor supply your own. For a controllable soundtrack use the Seedance family (audio parameter).
How is it billed, and am I charged on failure?
Per clip: the hold placed at submit is one clip's price and the settle is the exact same number. A failed or timed-out render refunds the whole hold, so you never pay for a clip you did not receive.

Related Models

Explore other models you can integrate.

View all →