Grok Video API (xAI)

Models and pricing

Both names are the same model and differ only in billing shape: grok-video-3 is per-second ($0.04/s at 480p / 720p, $0.12/s at 1080p), grok-video-3-10s is a flat per-clip price on six resolution × length tiers (10 seconds by default). Both do 1080p and output quality is identical, so take whichever is more affordable for your shape: **for 1080p at 6s or 10s use grok-video-3-10s** ($0.35 / $0.65 vs $0.72 / $1.20 per-second), and **for 1080p at 15s use grok-video-3** ($1.80 vs $2.00). 480p is only on grok-video-3, and 720p is more affordable on the per-second one.

Create a task

POST /api/v1/video/generations — async: create, then poll the same endpoint for the result.

# Step 1: Create task
curl -X POST https://api.apimodels.app/v1/video/generations \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "grok-video-3",
    "prompt": "A cat playing piano in a jazz club, cinematic lighting",
    "aspect_ratio": "16:9"
  }'

# Step 2: Poll status
curl "https://api.apimodels.app/v1/video/generations?task_id=TASK_ID" \
  -H "Authorization: Bearer YOUR_API_KEY"

Request parameters

Getting the result

The create call returns data.taskId right away and the video renders in the background. GET the same endpoint with that task_id and read data.state: pending means still running, completed puts the video URL in data.resultUrls[0], and failed puts the reason in data.failMsg. To skip polling entirely, pass callback_url and we POST you when the task finishes.

• Measured turnaround: a 6-second clip comes back in about 60-85 seconds; a 15-second one with several reference images and a long prompt can take up to 4 minutes. Set client timeouts to 6 minutes or more, or pass callback_url and skip polling entirely.
• Result videos are kept for 30 days and then deleted — download or re-host anything you want to keep.
• You are charged on success; failed tasks are not billed.

Things worth knowing

• Supports 16:9, 9:16, 1:1 and more, so landscape, portrait and square all come out of the same endpoint — no model switching for social formats.
• grok-video-3 is billed per second, with the rate set by resolution: 480p and 720p are both $0.04/s (6s $0.24, 10s $0.40, 15s $0.60), and 1080p is $0.12/s (6s $0.72, 10s $1.20, 15s $1.80). For short 1080p clips (6 / 10s), the flat tiers on grok-video-3-10s are more affordable.
• Image-to-video: put reference image URLs in images (up to 7) and the model follows their subject, setting and style. Omit it for pure text-to-video.
• grok-video-3-10s is billed per clip on flat tiers: 720p $0.30 / $0.50 / $0.70 and 1080p $0.35 / $0.65 / $2.00 for 6 / 10 / 15 seconds — the easier one to budget for batch jobs.
• Auth is the same as every other model: send Authorization: Bearer YOUR_API_KEY. One key covers every video, image, audio and LLM model on the platform.

Related