Animate images to 4K video with Kling 3 Omni. Combine an input image with up to 4 reference images in your prompt for ultra-high-resolution multi-image guided video generation. Try free — pay per generation, no subscription.
Kling v3 Omni at 4K. Multi-image reference video generation — supply up to 4 images and reference them in your prompt with <<<image_N>>>.
About this model
Kling v3 Omni at 4K (I2V). Multi-image reference video generation — reference up to 4 images in your prompt using <<<image_N>>>. Routes through apimart's managed Kling endpoint.
Cost analysis
| Provider | Cost | Notes |
|---|---|---|
| muapiapp | $0.536/sec base | Default 5-second clip costs $2.6785. Matches official Kling pricing. |
| Kling.ai (official) | Same | muapiapp routes to apimart's Kling provider; pricing mirrors Kling.ai's published rate. |
| Replicate / Fal.ai | Not available | Kling v3 Omni isn't generally available on Replicate or Fal.ai. |
Default 5-second clip costs $2.6785. Matches official Kling pricing.
muapiapp routes to apimart's Kling provider; pricing mirrors Kling.ai's published rate.
Kling v3 Omni isn't generally available on Replicate or Fal.ai.
** Competitor pricing is estimated based on similar model architectures and usage tiers.
Configuration schema
| Parameter | Type | Description | Default |
|---|---|---|---|
| Prompt | string | Text prompt. Reference images via <<<image_N>>> (1-indexed). If omitted, <<<image_1>>> is auto-prepended. | A high-speed train races forward nonstop while the environment transforms every few seconds—from snowy mountains to neon cyberpunk city to volcanic wasteland. Sparks fly from the tracks as the camera stays tightly locked alongside the speeding train during each violent world transition. |
| Image URLs | array | Up to 4 reference images. Address them in the prompt with <<<image_1>>> through <<<image_4>>>. | https://d3adwkbyhxyrtq.cloudfront.net/webassets/videomodels/kling-v3.0-omni-pro-image-to-video.jpg |
| Aspect Ratio | Enum (3 options) | Aspect ratio of the output video. | 16:9 |
| Duration | Enum (13 options) | Duration of the generated video in seconds. | 5 |
Text prompt. Reference images via <<<image_N>>> (1-indexed). If omitted, <<<image_1>>> is auto-prepended.
A high-speed train races forward nonstop while the environment transforms every few seconds—from snowy mountains to neon cyberpunk city to volcanic wasteland. Sparks fly from the tracks as the camera stays tightly locked alongside the speeding train during each violent world transition.Up to 4 reference images. Address them in the prompt with <<<image_1>>> through <<<image_4>>>.
https://d3adwkbyhxyrtq.cloudfront.net/webassets/videomodels/kling-v3.0-omni-pro-image-to-video.jpgAspect ratio of the output video.
16:9Duration of the generated video in seconds.
5Developer documentation
Upload reference images: Provide 1–4 images via images_list.
Reference them in the prompt: Use <<<image_1>>> … <<<image_4>>> to address each one. If you don't add any reference, <<<image_1>>> is auto-prepended.
Pick aspect ratio and duration: 16:9 / 9:16 / 1:1, 3–15 seconds.
Submit and poll: You'll receive a request_id immediately. Poll the result endpoint until status is completed.
Frequently asked
Use `<<<image_1>>>`, `<<<image_2>>>`, etc. — 1-indexed against the order you submitted in `images_list`. If you skip the references, the model auto-prepends `<<<image_1>>>` to your prompt.
No — at 4K, audio is bundled at no additional cost.
15 seconds. Minimum is 3 seconds.