Cookie Preferences
We use cookies to personalize content and analyze traffic. See our Cookie Policy.
Grok Imagine is xAI’s AI video generation model for creating high-quality videos with realistic motion, native audio, and synchronized speech. Grok Imagine 1.5 introduces faster generation speeds, improved motion physics, and stronger visual consistency for modern AI video applications.
No video yet. Submit the form to generate video.
Explore different use cases and parameter configurations
Complete guide to using grok-imagine
Seamlessly extend your existing AI-generated videos using xAI's grok-imagine model via the ApiPass integration platform. By providing a completed video task ID from a previous generation, you can describe how the scene should continue. The endpoint triggers a high-fidelity video-to-video (V2V) workflow, adding 6 or 10 seconds of seamless, context-aware new motion based on the final frame of your source clip.
Use ApiPass to extend an existing AI-generated video with the Grok Imagine Extend model ID. Provide a completed video task ID, describe how the scene should continue, and create an asynchronous video-to-video task that adds 6 or 10 seconds of new motion.
Continue a video from a completed ApiPass generation task. The input task_id must reference a previously successful video task; direct external video URLs are not supported for this endpoint.
Control how the clip continues with natural language. Describe camera movement, subject motion, lighting changes, environment transitions, and other temporal details to guide the extension.
Choose extend_times as 6 or 10 seconds. Shorter extensions are useful for quick continuations, while longer extensions give the model more time to develop the scene.
Create an extension task with POST /api/v1/jobs/createTask, store the returned taskId, and use a callback or GET /api/v1/jobs/recordInfo to retrieve completion status and final results.
Extend an existing generated clip while preserving the visual direction of the source task. This is useful when a strong first result needs a longer ending, an extra camera move, or a more complete story beat.
Use the prompt field to explain what happens next. You can specify camera pans, forward motion, subject actions, lighting changes, environmental details, and transitions so the continuation follows your creative intent.
The endpoint supports extend_times values of 6 and 10. This gives developers a simple way to balance output length, generation time, and iteration speed.
ApiPass returns a taskId after creation. In production, provide callBackUrl to receive final task updates automatically, or poll the recordInfo endpoint when you need to check status manually.
Turn a short generated clip into a more complete scene for social feeds, campaign assets, or short-form creative content by adding a continuation that matches the original motion and style.
Use extension prompts to explore what happens after a generated moment. Continue a character walk, reveal a new environment, or add a final cinematic beat without restarting the entire generation.
Extend product preview clips with extra camera movement, lighting changes, and hero-shot endings. This helps creative teams test multiple endings from a single successful source video task.
Integrate the asynchronous task flow into content tools, or internal pipelines. Generate an initial video, pass its taskId into grok-imagine/extend, and collect the final result through callback handling.
ApiPass offers a highly resilient infrastructure and abstraction layer to test, deploy, and scale enterprise-grade AI workflows through a unified developer dashboard.
Use text-to-video when you want to create a new video from a written prompt. Use Video Extend when you already have a completed ApiPass video task and want to add a continuation while building on the previous result. The recommended workflow is to create a video first, then pass its returned taskId as task_id for grok-imagine/extend.
This endpoint is designed for task-based continuation, not direct external video uploads. The task_id must come from a previously successful ApiPass video generation task. External video URLs are not supported for this model.
A 6-second extension is useful for fast iteration and short motion additions. A 10-second extension provides more room for a scene change or longer subject action, but it usually takes more time to generate.
The optional channel parameter can be set to auto, starter, regular, or official. The default auto channel lets ApiPass allocate tasks across available providers based on real-time pricing and stability metrics.
The endpoint supports 6-second and 10-second extensions. Longer extensions usually require more processing time, so choose the duration that best fits your creative and production needs.
New users can use free credits on ApiPass to test video generation workflows before moving into production. Use the playground to validate prompts, task IDs, and callback handling.
Start with a previously successful ApiPass video generation task. For example, create a video with grok-imagine/text-to-video and save the returned taskId. That value becomes the task_id input for grok-imagine/extend.
Describe how the video should continue. Include camera movement, subject actions, lighting changes, scene transitions, and any visual details that should carry the clip into the next moment.
Send a POST request to /api/v1/jobs/createTask with model set to grok-imagine/extend and input containing task_id, prompt, extend_at, and extend_times. Optionally include callBackUrl so ApiPass can send the final task result to your endpoint.
Store the returned taskId. If you supplied callBackUrl, listen for the completion callback. Otherwise, poll /api/v1/jobs/recordInfo?taskId=YOUR_TASK_ID to check status and retrieve the result when generation completes.
All APIs require authentication via Bearer Token.
Authorization: Bearer
Create an asynchronous video-to-video task that extends a previously generated Grok Imagine video.
1{
2 "model": "grok-imagine/extend",
3 "channel": "auto",
4 "callBackUrl": "https://your-domain.com/api/callback",
5 "input": {
6 "task_id": "task_grok_12345678",
7 "prompt": "The camera slowly pans forward, showing the protagonist walking deeper into the forest, with sunlight filtering through the leaves and casting dappled shadows.",
8 "extend_at": "0",
9 "extend_times": "6"
10 }
11}modelRequiredstringApiPass model endpoint name. Use grok-imagine/extend for this endpoint.
grok-imagine/extend
callBackUrlOptionalstringOptional callback URL. When provided, ApiPass sends the final task result to this URL after the task completes.
https://your-domain.com/api/callback
channelOptionalstringDefault channel is auto. APIPASS automatically allocates tasks across available providers based on real-time pricing and stability metrics; starter is ultra-low-cost with limited quotas and weaker stability; regular is standard and much cheaper than official APIs with moderate stability; official uses the model native API with high stability, fast task execution, and official pricing.
Default channel is auto. APIPASS automatically allocates tasks across available providers based on real-time pricing and stability metrics; starter is ultra-low-cost with limited quotas and weaker stability; regular is standard and much cheaper than official APIs with moderate stability; official uses the model native API with high stability, fast task execution, and official pricing.
Available options:
auto
task_idRequiredstringThe task ID of a previously and successfully generated video. It must come from a Kie AI video generation model, such as grok-imagine/text-to-video. Only Kie AI-generated videos are supported; external video URLs are not supported.
Maximum length is 100 characters. The original video task must already be completed successfully.
task_grok_12345678
promptRequiredstringDescribes how the video should continue during the extension, including camera movement, scene changes, subject actions, and other motion details. More specific prompts usually produce results that better match expectations. Chinese and English prompts are supported.
The camera slowly pans forward, showing the protagonist walking deeper into the forest, with sunlight filtering through the leaves and casting dappled shadows.
extend_atRequiredstringThe starting point of the video extension. Although the upstream parameter description may mark it as optional, the required list and current ApiPass adapter require this field.
0
extend_timesRequiredstringThe duration of the video extension in seconds. 6 extends the video by 6 seconds, and 10 extends the video by 10 seconds. Longer extensions usually take more time to generate.
Available options:
6
1curl -X POST "https://api.apipass.dev/api/v1/jobs/createTask" \
2 -H "Authorization: Bearer YOUR_API_KEY" \
3 -H "Content-Type: application/json" \
4 -d '{
5 "model": "grok-imagine/extend",
6 "channel": "auto",
7 "callBackUrl": "https://your-domain.com/api/callback",
8 "input": {
9 "task_id": "task_grok_12345678",
10 "prompt": "The camera slowly pans forward, showing the protagonist walking deeper into the forest, with sunlight filtering through the leaves and casting dappled shadows.",
11 "extend_at": "0",
12 "extend_times": "6"
13 }
14}'1{
2 "taskId": "task_xxx",
3 "status": "queued"
4}taskIdThe task ID returned by ApiPass. Use this value to query task status or match callback results.
statusThe initial task status after creation, such as queued.
Google Veo 3.1 is the latest state-of-the-art generative video model designed to transform text and image prompts into high-fidelity cinematic visuals. Building upon its predecessors, Veo 3.1 features significant upgrades in prompt adherence, visual realism, and native audio generation, creating 8-second clips with synchronized sound effects and dialogue.
Starting from
0 credits
topaz
topaz
Topaz Video Upscaler API delivers professional-grade AI video enhancement, restoring detail, reducing noise, and providing high-quality upscaling to 1080p or 4K.
Starting from
0 credits
Luma
Luma
Starting from
0 credits
Starting from
0 credits