← Home

Guide

Grok Imagine Video on OpenClaw: xAI Video for Your Agent

Grok Imagine Video is the cheapest way to give an agent video, starting at $0.05 per second with audio included. It is a different product from the Grok chat models — it lives in your instance's Video slot, not the Model slot. Here is how to wire it up and where v1.5 is worth the premium.

Grok Imagine Is Not the Grok You Chat With

Worth clearing up first, because the naming invites confusion. The Grok models covered in our Grok API setup guide are chat models that go in your instance's Model slot. Grok Imagine Video is a generation model that goes in the Video slot. You can run both at once — a Grok chat brain that produces Grok-generated clips — or mix freely, since the two slots are independent.

Two Routes

  • xAI key — add an xAI key on your API Keys page and Grok Imagine Video and Grok Imagine Video 1.5 appear as curated rows in the xAI group of the Video dropdown.
  • fal key — the same models are also served through fal.ai, which is convenient if you already keep Kling, Seedance and MiniMax H3 on one bill.

v1 vs v1.5

Both generate video with audio. v1 is the budget option and caps out at 720p; v1.5 adds 1080p and costs meaningfully more at every resolution.

ResolutionGrok Imagine VideoGrok Imagine Video 1.5
480p$0.05 / sec$0.08 / sec
720p$0.07 / sec$0.14 / sec
1080p$0.25 / sec

On v1, image input adds a flat $0.002, so a 6-second 480p image-to-video clip lands at about $0.302 and the text-to-video equivalent at $0.30. On v1.5 a 5-second clip is $0.40 at 480p, $0.70 at 720p and $1.25 at 1080p, scaling linearly with duration.

This is the volume option. At $0.05 per second, Grok Imagine v1 is roughly a sixth the price of Seedance 2.0 standard and a quarter the price of Veo 3.1 per second of output. If you want an agent that can answer “make me a video” in a busy group chat without the bill becoming the story, this is where to start — and step up to v1.5 or another family only for work that ships.

Setting It Up

  1. Add an xAI key (from console.x.ai) or a fal key (from fal.ai/dashboard/keys) on your API Keys page.
  2. On your instance card, open the Video dropdown and select the xAI or fal.ai provider group.
  3. Pick Grok Imagine Video or Grok Imagine Video 1.5. On the fal side the visible list refreshes daily, so type “Grok” in the dropdown search box if you do not see it.
  4. Ask your agent for a clip in chat and the MP4 comes back in the conversation.

Model IDs for Self-Hosted OpenClaw

xAI route:

xai/grok-imagine-video
xai/grok-imagine-video-1.5

fal route (fal endpoint path with a fal/ prefix):

fal/xai/grok-imagine-video/text-to-video
fal/xai/grok-imagine-video/image-to-video
fal/xai/grok-imagine-video/v1.5/image-to-video

What's Next?

Give Your Agent a Video Model

Deploy an always-on agent, add your xAI or fal key, and pick Grok Imagine Video.

Deploy Now