Access Grok Imagine Video 1.5 (xAI's Grok video model) through EvoLink's unified video API. Test text-, image-, and reference-to-video with native audio first.
Write a prompt, optionally attach up to 7 images, estimate credits, and preview the output before you wire up the async API.
Create a Sample
Add a prompt, attach images if you need them, and set the parameters.Optional, up to 7 images. 0 images = text-to-video · 1 = image-to-video · 2-7 = reference-to-video
1generated-image.png
—
✓ 1 image(s) added
Not applied with one input image — the output follows that image's dimensions.
Results are kept for 24 hours — download anything you want to keep.
Plan credits for every mode. Output is billed by duration and resolution; each input image adds a small fixed fee, so prompt-only text-to-video carries no image charge.
First Test Cost
Text-to-Video - 1s - 480pOne 1s 480p sample without input images
About 178 tests with $10 in credits.
Iterate in 480p to save credits, then switch to 720p or 1080p when the motion and framing are ready for review.
Testing Budget Guide
Pick an amount based on how many tests you expect.Good for a first validation.
Good for iterating on motion and framing.
Good before production integration.
Estimated total for one run with 1 input image. Final cost may vary by duration, resolution, and how many images you attach.
grok-imagine-video-1.5-preview| Item | 480p | 720p | 1080p | Billing |
|---|---|---|---|---|
| Video generationOutput video, per second | $0.056/s-30% 3.808 cr/s$0.080 Official price | $0.098/s-30% 6.664 cr/s$0.140 Official price | $0.175/s-30% 11.9 cr/s$0.250 Official price | Output video seconds |
Output is billed per second (720p = 480p x 1.75, 1080p = 480p x 3.125). Each input image adds a small fixed fee, charged per image from 0 to 7.
| Add-on | When it applies | Price | Credits |
|---|---|---|---|
| Input image | Per input image (0-7, optional) | $0.0070/image-30% $0.010 Official price | 0.476 cr |
Input images are optional. Each attached image is billed once at a fixed rate across all plans, and a prompt-only text-to-video request has no image charge.
Grok Imagine Video 1.5 is xAI's GA video model. On EvoLink, one model ID covers three generation modes and the number of input images selects between them: send none for text-to-video, one to animate a source image, or two to seven to blend reference images. Every request takes a required prompt and returns a 1-15 second video at 480p, 720p, or 1080p. The route is designed for short-form generation with native audio, asynchronous task handling, live cost estimation, and one API key shared with other models.
grok-imagine-video-1.5-previewOne model ID covers all three modes - no image for text-to-video, one image to animate it, or 2-7 to blend references. xAI's upstream stable ID is grok-imagine-video-1.5; EvoLink currently routes the preview alias.
Use the EvoLink API for product integration and batch jobs, or call it from Codex, Claude, or Gemini for fast creative and development workflows. Both paths share the same EvoLink API key, balance, model routes, and task history.
Best for: product backends, batch jobs, automated workflows
Call EvoLink’s unified video API from your server and control the model ID, parameters, task queue, callbacks, and result storage.
Best for: creative and development tasks in Codex, Claude, and Gemini
Give the Agent your output goal, assets, and acceptance criteria. It can choose the route, assemble the request, track the task, and return the result without requiring you to hand-code every step.
Generate from a prompt alone, animate a prepared product shot or storyboard frame, or blend two to seven reference images - without switching model names or endpoints.
The model can generate motion with synchronized ambience, sound effects, and speech in the same pass. Validate audio quality against your own acceptance criteria.
Submit once, track the task by polling or callback, then persist the completed asset before its temporary result URL expires.
Turn product photography and campaign key visuals into short motion concepts and creative variants.
Try in PlaygroundCreate portrait, square, or landscape clips with camera direction and optional audio cues for social workflows.
Try in PlaygroundExplore camera movement, character performance, physical effects, timing, and scene rhythm from a storyboard frame before committing to a larger production.
Try in PlaygroundThis example shows the shortest runnable flow: create a task, save the returned task ID, then poll or wait for an HTTPS callback. Open the API tab for the complete parameter and response reference.
curl -X POST https://api.evolink.ai/v1/videos/generations \
-H "Authorization: Bearer $EVOLINK_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "grok-imagine-video-1.5-preview",
"prompt": "The person starts dancing under stage lights",
"duration": 6,
"quality": "480p",
"aspect_ratio": "16:9",
"callback_url": "https://your-domain.com/webhooks/video-task-completed"
}'
# Save the returned task ID, then query:
curl https://api.evolink.ai/v1/tasks/{task_id} \
-H "Authorization: Bearer $EVOLINK_API_KEY"Check every value against the parameter cards above: ranges, allowed options, asset counts, and file types must match the selected route.
Check the Authorization bearer token and confirm the available balance and task billing in the console.
Review likeness rights, brand assets, readable text, sensitive content, and input-asset compliance.
A Grok Imagine Video 1.5 task that reaches a final failed state is not charged. Keep the task ID, inspect the final state, and identify whether the cause is validation, review, or execution before retrying.
callback_url must be a public HTTPS URL, not localhost or a private network address. Always keep the task ID as a polling fallback.
| Feature | Grok Imagine Video 1.5 | Production impact | Choose another route when |
|---|---|---|---|
| Generation mode | Text, image, and reference-to-video | The image count picks the mode, so one integration covers all three. | You need first-and-last-frame or video-to-video editing. |
| Input contract | 0-7 images + non-empty prompt | Validate the prompt and cap the image list at 7 before submitting. | You need more than 7 references or a video input. |
| Duration & output | 1-15s at 480p, 720p, or 1080p | Cost scales with duration and resolution. | You require clips longer than 15 seconds on EvoLink. |
| Output framing | auto by default, or 5 fixed ratios; a single source image always wins | Set the ratio for prompt-only and multi-reference runs; auto hands framing to the upstream channel. | You need 4:3, 3:4, or any ratio outside the supported set. |
| Delivery | Async task with polling or callback | Queue tasks and make terminal handling idempotent. | Your workflow requires an immediate synchronous result. |
| Billing | Per output second + per input image | Estimate each task and cap retry spending. | You need a fixed cost independent of duration. |
xAI lists grok-imagine-video-1.5 as the stable model. EvoLink currently accepts grok-imagine-video-1.5-preview, so use the ID shown in EvoLink examples.
There is no mode parameter. Omit image_urls for text-to-video, send one JPEG, PNG, or WebP image up to 10 MB to animate it as the first frame, or send two to seven images to blend them as references. Eight or more images are rejected.
Submit a task, poll or receive a callback, and persist the finished video. EvoLink result URLs are available for 24 hours.
Use live route pricing to estimate duration, resolution, and input-image cost before generation.
Call Grok Imagine and other video models through one authentication and integration layer.
Manage credits, spending, and model usage in the same EvoLink account.
Track pending, processing, completed, and failed tasks through a consistent async workflow.
Keep Grok 1.5 for prompt, single-frame, and multi-reference generation, and route other modalities to a better-matched model.
Compare Grok Imagine routes under one EvoLink account and API key, then choose by output quality, duration, and cost.
Grok Imagine Video 1.5
Grok Imagine Video 1.5 is xAI's GA video model. On EvoLink, one model ID covers three generation modes and the number of input images selects between them: send none for text-to-video, one to animate a source image, or two to seven to blend reference images. Every request takes a required prompt and returns a 1-15 second video at 480p, 720p, or 1080p. The route is designed for short-form generation with native audio, asynchronous task handling, live cost estimation, and one API key shared with other models.
View API
Grok Imagine Video (1.0)
The broader Grok Imagine video route for text-to-video, image-to-video, reference, edit, and extend workflows.
View API
Seedance 2.0
A video model for text, image, and reference-driven generation with native audio.
View API
Grok Imagine Video 1.5 is xAI's GA video model. It turns a prompt, one source image, or two to seven reference images into a short video with native audio capabilities.
Use grok-imagine-video-1.5-preview in EvoLink requests. xAI's upstream stable ID is grok-imagine-video-1.5, but model IDs are channel-specific.
The preview name remains an upstream alias and is the current public ID for the EvoLink route. It does not mean xAI still classifies the 1.5 release as Preview.
Yes. Omit image_urls, or send an empty array, and the request runs as text-to-video from the prompt alone. No image fee is charged in that case.
Yes. Send two to seven images in image_urls and the request runs as reference-to-video, blending them together. Eight or more images are rejected. A first-and-last-frame workflow is still not supported.
You do not pass a mode parameter. The number of valid images in image_urls decides it: zero runs text-to-video, one runs image-to-video with that image as the opening frame, and two to seven run reference-to-video, blending them together.
The prompt is always required and can be up to 2,000 tokens (CJK text uses more tokens per character). Images are optional: send up to seven JPEG, PNG, or WebP files of 10 MB or less.
EvoLink bills output by duration and resolution and adds a per-image charge for each attached input image. Use the live pricing section and playground estimate for the current route price instead of relying on a copied static figure.
The EvoLink route supports 480p, 720p, and 1080p output with durations from 1 to 15 seconds. Output is billed per second from the 480p rate: 720p costs 1.75x and 1080p costs 3.125x, matching xAI's own per-second pricing.
Text-to-video and reference-to-video (2-7 images) accept an aspect_ratio parameter: auto (default), 16:9, 9:16, 1:1, 3:2, or 2:3. auto lets the upstream channel decide the framing, so it can differ between runs — specify a fixed ratio when you need a predictable frame. When you attach a single input image, the parameter is ignored and the output follows that image's dimensions, so prepare it in the framing you want.
The model is designed to generate synchronized ambience, sound effects, and speech with the video. Audio quality can vary by scene, so test it against your own acceptance criteria.
Generation is asynchronous. Submit a task, poll its status or provide an HTTPS callback, and persist the returned video when the task completes.
EvoLink result URLs are available for 24 hours. Copy completed assets to your own object storage instead of using the temporary URL in production.
Do not assume that every failed or moderated request is free. Inspect the task's returned usage and billing fields and confirm the current account record before retrying.