Grok Imagine Video 1.5 Preview: Text-to-Video, Image-to-Video, and the CodeFast Grok Imagine Guide
Grok Imagine Video 1.5 Preview stands out in xAI's short-video generation stack with text and image input, 1-15 second duration, 480p/720p options, and per-second pricing. CodeFast Grok Imagine Studio makes the same workflow easier to test through packages, daily credits, and one-panel management.
· CodeFast Team
What is Grok Imagine Video 1.5 Preview?
As of June 1, 2026, xAI's model documentation lists Grok Imagine Video 1.5 Preview under the official model name grok-imagine-video-1.5-preview. Its alias is grok-imagine-video-1.5-2026-05-30. The model accepts text and image input and produces video output, so it is positioned for prompt-based video generation and turning an existing image into a short clip.
The important distinction is that the Grok Imagine ecosystem contains separate flows for image generation, image editing, text-to-video, and image-to-video. For 1.5 Preview, the clearest value is better short-video generation and image-to-video workflows that can be controlled from an API.
Official technical and pricing snapshot
xAI's model page lists Grok Imagine Video 1.5 Preview output pricing at $0.08 per second for 480p and $0.14 per second for 720p. When image input is used, image input is listed at $0.01. The model page also shows us-east-1 and eu-west-1 regions plus a 60 requests-per-minute limit.
Official xAI snapshot - June 1, 2026
Model: grok-imagine-video-1.5-preview
Alias: grok-imagine-video-1.5-2026-05-30
Input: text, image
Output: video
Duration: 1-15 seconds
Resolution: 480p or 720p
Output price: $0.08/sec at 480p, $0.14/sec at 720p
Image input: $0.01
Regions: us-east-1, eu-west-1
Rate limit: 60 requests per minute
Official prices and limits can change; check the xAI model and pricing pages again before production use.
How the video generation flow works
xAI's video generation documentation uses an asynchronous flow: start a generation request, then poll with the request_id until the result is ready. When complete, the response returns a video URL. Because the hosted URL can be temporary, production apps should download and store the video in their own storage layer.
- Duration: configurable from 1 to 15 seconds.
- Aspect ratio: options such as 16:9, 9:16, 1:1, 4:3, 3:4, 3:2, and 2:3 provide enough flexibility for short-video use cases.
- Resolution: 480p makes sense for faster and cheaper tests, while 720p is better for cleaner presentations.
- Mode: prompt alone creates text-to-video; prompt plus image creates an image-to-video flow.
Where CodeFast Grok Imagine Studio helps
Direct xAI API gives the most control, but not every team wants to manage per-second billing, a separate API key, separate storage flow, and separate limit tracking during early tests. CodeFast Grok Imagine Studio adds a package layer: a 30-day package, 100 daily Grok Imagine credits by default, 1 credit for image generation, 2 credits for image editing, and 5 credits for video generation.
That matters for idea validation, customer demos, campaign creative, and application prototypes. Testing the short-video model through a CodeFast package first gives teams a calmer way to judge prompt quality, output ratio, and product flow before moving to a deeper provider integration.
curl https://api.codefast.app/grok-imagine-api/v1/video/generate \
-H "Authorization: Bearer cf_live_your_api_key" \
-H "Content-Type: application/json" \
-d '{
"model": "grok-imagine-video-1.5-preview",
"prompt": "A clean product demo shot, slow camera push, premium lighting, smooth motion",
"aspect_ratio": "16:9",
"video_length": 8,
"resolution": "720p"
}'
A short text-to-video example for CodeFast Grok Imagine API. For image-to-video, add an image_url field.
Best use cases
- SaaS product demos: turning a single screenshot into a short onboarding or feature teaser clip.
- E-commerce creative: turning product photos into animated ad variations.
- Mobile social content: creating fast 9:16 story, shorts, or vertical campaign drafts.
- Game and app concepts: extracting short motion, camera, and atmosphere tests from a static concept image.
- Agency workflow: using the same prompt structure to create fast variations for different clients, products, and formats.
How to write better prompts
A video prompt needs more time information than an image prompt. It is not enough to describe the scene; you should describe motion, camera, pace, lighting, and the beginning-to-end state. Short-video models usually work best with one main idea, one camera movement, and a clear sense of duration.
- Choose the format first: 16:9 demo, 9:16 social clip, or 1:1 product showcase.
- Define the motion in one sentence: slow push-in, orbit, pan, reveal, or parallax.
- Constrain the scene: one product, one background, one lighting style, minimal camera changes.
- Do not overdo negative instructions: write the desired result clearly instead of trapping the model in a long ban list.
Grok Imagine 1.5, Veo, or Omni Flash?
There is no single correct answer. Grok Imagine 1.5 is a good option for fast iteration on short-video and image-to-video tests. Veo and Omni Flash may fit better when you need Google video workflows, longer generation strategies, or an already Google-centered stack. The useful CodeFast angle is that Grok Imagine, Veo, Omni Flash, and image-generation packages can be compared from the same panel logic.
Simple cost math
With direct xAI API pricing, an 8-second 480p video costs about $0.64 in output. The same duration at 720p costs about $1.12. If image input is used, the official table also lists a $0.01 image input cost. These calculations help estimate the provider-side cost per video.
CodeFast works differently: you buy a package, see a daily credit allowance, and use Studio plus API flows through the same rights. With the default 100 daily credits and 5 credits per video, the base package can be planned around roughly 20 video tests per day if you only generate video. This can change based on selected limits, duration, and current dashboard rules.
Checklist before production
- Do not treat output URLs as permanent; move results into your own storage layer.
- Log prompts and settings; track which aspect ratio, duration, and resolution produce good results.
- Design queue and retry behavior; video generation may not return instantly like a text response.
- Set credit or budget limits; 720p and longer tests can create cost quickly.
- Keep human review for brand safety; outputs should be reviewed before ads, product pages, or social publishing.