# Grok Imagine Video: clip length, audio, price and limits

> Grok Imagine Video 1.5 makes 1-15 second clips with native audio from $0.08 a second at 480p; Lite is $0.02. Specs and per-clip costs vs Veo 3.1 and Kling.

URL: https://honeyvid.ai/learn/grok-imagine-video/
Updated: 2026-10-30
Publisher: HoneyVid (https://honeyvid.ai)
Topic: Apps and pricing

## Short answer

Grok Imagine Video is xAI's text- and image-to-video model, with sound and dialogue generated in the same pass. The API makes 1 to 15 second clips and charges from $0.08 a second for 1.5 and $0.02 for 1.5 Lite at 480p, rising with resolution (xAI pricing, 2026-10-26). For the cheapest tries, use Lite; for quality or voice control, use 1.5.

## Key facts

- **Grok Imagine Video 1.5 announced:** 16 June 2026, generally available in the xAI API ([source](https://x.ai/news/grok-imagine-video-1-5))
- **API output price per second at 480p / 720p / 1080p (as of 2026-10-26):** 1.5: $0.08 / $0.14 / $0.25. 1.5 Lite: $0.02 / $0.03 / $0.14. Classic grok-imagine-video: $0.05 / $0.07 / no 1080p ([source](https://docs.x.ai/developers/pricing))
- **Clip length:** 1 to 15 seconds, default 8 ([source](https://docs.x.ai/developers/model-capabilities/video/generation))
- **Resolution:** 480p default; 1.5 native 1080p, Lite 1080p upscaled from 720p, classic up to 720p ([source](https://docs.x.ai/developers/model-capabilities/video/generation))
- **Audio:** On by default with lip-synced speech on 1.5 and Lite; up to 3 preset voices via reference audio on 1.5 ([source](https://docs.x.ai/developers/model-capabilities/video/generation))
- **Rate limit, 1.5:** 10 requests per second ([source](https://docs.x.ai/developers/models/grok-imagine-video-1.5))
- **Veo 3.1 with audio on Runway's API:** $0.40 per second; 4, 6 or 8 second clips ([source](https://docs.dev.runwayml.com/guides/pricing/))

This guide compares xAI's documented specifications and API prices, checked on 2026-10-26, and turns them into the cost of a clip. Where xAI makes a quality claim, the page says it is xAI's claim.

## What is Grok Imagine Video?

Grok Imagine Video is xAI's video model: it turns a text prompt or a starting image into a short clip with sound.

xAI [announced Grok Imagine Video 1.5](https://x.ai/news/grok-imagine-video-1-5) on 16 June 2026 with the line "Improved quality at even faster speeds". The model went from preview to generally available in the xAI API as `grok-imagine-video-1.5`. A faster variant, Video 1.5 Fast, rolled out in the Grok web app and the iOS and Android apps. xAI says Fast makes a 6-second 720p video in about 25 seconds, against 40 seconds or more for the previous model. That is a speed claim from xAI, not a measurement.

Three video models are in the API as of 2026-10-26, per [xAI's models page](https://docs.x.ai/developers/models). Prices below are output per second at 480p, the default, from [xAI's pricing page](https://docs.x.ai/developers/pricing):

| Model | Price per second (480p) | Resolution | Notes |
| --- | --- | --- | --- |
| `grok-imagine-video-1.5` | $0.08 | Native 1080p (reference-to-video up to 720p) | Recommended for quality. Reference images, voice references, pinned first, last or key frames |
| `grok-imagine-video-1.5-lite` | $0.02 | 1080p upscaled from 720p | Lowest price. Recommended for simple videos |
| `grok-imagine-video` (classic) | $0.05 | Up to 720p | Supports edit and extend, no last frame or key frames |

## How long can a Grok Imagine video be?

A Grok Imagine API clip can be 1 to 15 seconds long, with 8 seconds as the default.

Other limits from [xAI's video guide](https://docs.x.ai/developers/model-capabilities/video/generation):

- Reference-to-video on the classic model is capped at 10 seconds.
- Editing a clip keeps the original length, capped at 8.7 seconds, and output is capped at 720p.
- Resolution defaults to 480p. You must ask for 720p or 1080p.
- Aspect ratios include 16:9 (default), 9:16, 1:1, 4:3, 3:4, 3:2, 2:3, 21:9 and 5:2.
- Image-to-video follows the shape of your input image and ignores the aspect ratio setting.

Fifteen seconds is almost double the 8 seconds Google lists for Veo 3.1 in the [Gemini API docs](https://ai.google.dev/gemini-api/docs/veo). Length is the cheap part. A longer clip leaves more room for drift, so judge quality on your own 8-second and 15-second runs.

## Does Grok Imagine generate audio and dialogue?

Grok Imagine generates audio with the video: sound effects, ambience and dialogue come out of the same pass.

Audio is on by default. Set `generate_audio=False` for a silent clip. Lip-synced speech is listed for both 1.5 and 1.5 Lite. On 1.5, reference-to-video can assign a preset voice to up to 3 speakers through `reference_audios`, and audio input for preset voices is free. xAI says speech is clearer and better synced than in the earlier model. It publishes no sync score, so that is a vendor claim. Test one line of dialogue on your own subject before you build a series on it.

xAI lists no separate audio charge, unlike Runway's listing for Veo and Kling (more in [Runway pricing](https://honeyvid.ai/learn/runway-pricing/)).

## How much does Grok Imagine cost?

Grok Imagine costs $0.08 a second on 1.5, $0.05 on the classic model and $0.02 on 1.5 Lite at the default 480p in the xAI API, and the rate rises with resolution.

| Model | 480p | 720p | 1080p |
| --- | --- | --- | --- |
| 1.5 Lite | $0.02/s | $0.03/s | $0.14/s |
| Classic | $0.05/s | $0.07/s | not offered |
| 1.5 | $0.08/s | $0.14/s | $0.25/s |

Input is billed too, per [xAI's pricing](https://docs.x.ai/developers/pricing): $0.01 per input image on 1.5 and Lite, $0.002 per image on classic, and $0.01 per second of input video on classic. An image-to-video clip on 1.5 therefore costs the output seconds plus $0.01.

For an 8-second clip, output only:

| Model | 480p | 720p | 1080p |
| --- | --- | --- | --- |
| 1.5 Lite | $0.16 | $0.24 | $1.12 |
| Classic | $0.40 | $0.56 | n/a |
| 1.5 | $0.64 | $1.12 | $2.00 |

The Grok app uses plans, not per-second billing. xAI's developer docs list no app caps or subscription prices.

## Grok Imagine vs Veo 3.1 and Kling: cost per clip

At list price, an 8-second Grok Imagine 1.5 clip at 480p is $0.64, against $3.20 for Veo 3.1 with audio and $1.04 for Kling 3.0 Standard with audio.

Veo and Kling rates below are from [Runway's API price list](https://docs.dev.runwayml.com/guides/pricing/), where one credit is $0.01. Runway's API takes 4, 6 or 8 seconds for Veo, so the comparison uses 8 seconds. Grok rates are xAI's own, output only. Both are API list prices on 2026-10-26, so they compare on price, not quality.

| Model | Per second (with audio) | 8-second clip | Three tries per keeper |
| --- | --- | --- | --- |
| Grok Imagine 1.5 Lite, 480p | $0.02 | $0.16 | $0.48 |
| Grok Imagine 1.5, 480p | $0.08 | $0.64 | $1.92 |
| Kling 3.0 Standard (Runway) | $0.13 | $1.04 | $3.12 |
| Grok Imagine 1.5, 720p | $0.14 | $1.12 | $3.36 |
| Veo 3.1 Fast (Runway) | $0.15 | $1.20 | $3.60 |
| Kling 3.0 Pro (Runway) | $0.17 | $1.36 | $4.08 |
| Grok Imagine 1.5, 1080p | $0.25 | $2.00 | $6.00 |
| Veo 3.1 (Runway) | $0.40 | $3.20 | $9.60 |

The last column multiplies the clip price by 3. The 3 is an assumption, not a measured hit rate. At one usable take in three, a kept 8-second clip costs $9.60 on Veo 3.1 and $1.92 on Grok Imagine 1.5 at 480p, a five-to-one gap. At 1080p the gap falls to $6.00 against $9.60. If Grok needs six tries at 480p and Veo needs two, Grok costs $3.84 and Veo $6.40. Count your own keeper rate on a small batch first.

Runway's Veo 3.1 rate halves to $0.20 a second without audio, which narrows the gap for silent clips.

The Veo 3.1 preview models in the Gemini API were shut down on 22 October 2026, and Google's docs point to Gemini Omni Flash and its enterprise platform instead. Prices for Veo through Google may differ from Runway's; see [Google Veo 3 pricing](https://honeyvid.ai/learn/google-veo-3-pricing/). For a fuller table of Kling plans and credits, see [Kling AI pricing](https://honeyvid.ai/learn/kling-ai-pricing/), and for a quick sum on your own clip lengths use the [AI video cost calculator](https://honeyvid.ai/learn/ai-video-cost-calculator/).

## Grok Imagine vs Sora

Sora is gone: OpenAI closed the Sora web and app on 26 April 2026, and the Sora 2 API [shut down on 24 September 2026](https://developers.openai.com/api/docs/deprecations) with no replacement named.

[Sora alternatives](https://honeyvid.ai/learn/sora-alternatives/) lists the live options with their dates and costs. At 480p, Grok Imagine Lite has the lowest output price of the models compared in the table above. Against Sora's old strengths, Grok Imagine offers audio with dialogue, image-to-video and a 15-second maximum.

## What is Grok Imagine Lite?

Grok Imagine Lite is `grok-imagine-video-1.5-lite`, the lowest-priced Grok video model at $0.02 per second at 480p. It sits among the other [apps and pricing guides](https://honeyvid.ai/learn/topics/platforms/).

xAI's guide says it is for simple videos and gives it lip-synced audio like 1.5. Its 1080p output is upscaled from 720p, while 1.5 renders native 1080p. Reference voices and reference-image video are documented for 1.5, not Lite. Use Lite to try prompts, framing and timing, then rerun a keeper on 1.5. The two models are different, so a Lite result will not repeat exactly on 1.5.

## What can't Grok Imagine do?

Grok Imagine has documented limits on resolution for some modes, clip editing and reference inputs.

- Reference-to-video on 1.5 tops out at 720p.
- The classic model edits and extends clips. 1.5 does not list those endpoints.
- The classic model cannot combine an image with reference inputs.
- [Reference-to-video](https://docs.x.ai/developers/model-capabilities/video/reference-to-video) accepts up to 14 reference images on 1.5 and up to 7 on classic.
- 1.5 allows [10 requests per second](https://docs.x.ai/developers/models/grok-imagine-video-1.5).

## Can you use real faces and characters?

Using a real person's face or a copyrighted character in a Grok Imagine video falls under xAI's usage policy and the rights of the person or owner.

Get consent before you make a video of a real person, and publish a clip with a known character only if you hold the rights.

## A checklist before you pick a Grok model

1. Need cheap drafts? Use Lite at $0.02 per second.
2. Need native 1080p, reference images or chosen voices? Use 1.5.
3. Need to edit or extend an existing clip? Use the classic model.
4. Need dialogue? Generate one short line first and check the lip sync.
5. Count your keeper rate on five clips, then multiply by the per-second rate at your resolution.

## Questions

### How long can a Grok Imagine video be?

In the xAI API a clip can run from 1 to 15 seconds, with 8 seconds as the default. Reference-to-video on the classic model stops at 10 seconds, and edits keep the source length up to 8.7 seconds. The Grok app's limits are not in xAI's developer docs.

### Is Grok Imagine free?

xAI's API is paid per second of output. Free or capped use in the Grok app depends on your plan, and xAI's developer docs list no app caps.

### What is Grok Imagine Lite?

Lite is grok-imagine-video-1.5-lite, the lowest-priced model at $0.02 per second at 480p ($0.03 at 720p, $0.14 at 1080p). xAI recommends it for simple videos. It keeps lip-synced audio but renders 1080p by upscaling 720p, and xAI documents reference voices for 1.5 only.

### Does Grok Imagine make audio and dialogue?

Yes. xAI says sound effects, ambience and dialogue are generated with the video and aligned to the action. Audio is on by default and you can switch it off with generate_audio set to false. Check lip sync on one short line before you batch.

### Is there a Grok Imagine API?

Yes. grok-imagine-video-1.5 left preview and became generally available on 16 June 2026. It takes a prompt or a starting image and lets you set duration, resolution and aspect ratio. Editing and extending clips use the classic model through separate endpoints.

## Sources

- [Grok Imagine Video 1.5](https://x.ai/news/grok-imagine-video-1-5), xAI (accessed 2026-10-26)
- [Models and pricing](https://docs.x.ai/developers/models), xAI Docs (accessed 2026-10-26)
- [Video generation](https://docs.x.ai/developers/model-capabilities/video/generation), xAI Docs (accessed 2026-10-26)
- [grok-imagine-video-1.5](https://docs.x.ai/developers/models/grok-imagine-video-1.5), xAI Docs (accessed 2026-10-26)
- [Pricing](https://docs.x.ai/developers/pricing), xAI Docs (accessed 2026-10-26)
- [Reference to video](https://docs.x.ai/developers/model-capabilities/video/reference-to-video), xAI Docs (accessed 2026-10-27)
- [Deprecations](https://developers.openai.com/api/docs/deprecations), OpenAI (accessed 2026-10-27)
- [API pricing](https://docs.dev.runwayml.com/guides/pricing/), Runway (accessed 2026-10-26)
- [Generate videos with Veo](https://ai.google.dev/gemini-api/docs/veo), Google AI for Developers (accessed 2026-10-26)