img_uuid, then poll or receive a callback. GoEnhance provides independent access and is not Kuaishou or Kling’s official API, partner, or model developer.Kling O1 API for Multimodal Video Generation
Access Kling O1 through GoEnhance for prompt-only video, first-frame or first-and-last-frame control, ordered reference images, multi-image elements, and one reference video. Submit one asynchronous request, save the returned
Get API Key
One endpoint for several reference-driven workflows
Prompt, frame, and reference control
POST to https://api.goenhance.ai/api/v1/videos/generations with model: kling-o1. A prompt is required. Add a first frame, a first-and-last-frame pair, ordered ref_imgs, element IDs, or one ref_videos input when the material rules allow it.
Generate from motion or edit a base video
With a reference video, use video_type: feature to imitate motion or style, or video_type: base to edit the supplied video in place. O1 can retain that video’s original sound, but it cannot generate a new audio track.
Validate material combinations before submitting a task.
Kling O1 API capabilities and validation rules
Model and prompt
Set <code>model</code> to <code>kling-o1</code>. The prompt is required and accepts up to 2,500 characters; use indexed references such as <code>@image_1</code>, <code>@element_1</code>, and <code>@video_1</code>.
Frames and references
Use <code>image_url</code> for the first frame and <code>image_end_url</code> for the last frame. Reference images use ordered <code>ref_imgs</code>; one reference video may be supplied through <code>ref_videos</code>.
Material budget
Frames, reference images, and element IDs may total no more than 7 materials, or no more than 4 when a reference video is used. A first-and-last-frame pair disallows extra reference images and element IDs.
Duration
Choose an integer from 3 to 10 seconds; the documented default is 5. A first-frame request without reference images or video accepts only 5 or 10 seconds.
Resolution and ratio
Choose 720p or 1080p. Text-only requests accept 16:9, 9:16, or 1:1; a first frame or reference video determines the ratio instead. O1 has no 4K tier.
Audio behavior
<code>generate_audio</code> does not create new audio. With a reference video, it can retain the original sound; otherwise O1 produces no generated audio track.
GoEnhance bills generated video per second at $0.02 per token.
Kling O1 API pricing
720p
Without video input: 4.2 tokens per second ($0.084/s). With video input: 6.3 tokens per second ($0.126/s).
1080p
Without video input: 5.6 tokens per second ($0.112/s). With video input: 8.4 tokens per second ($0.168/s).
Estimate each task
Multiply the requested duration by the exact rate for the selected resolution and whether a reference video is present. Retaining reference-video audio has no separate surcharge.
Integrate Kling O1 in six steps
1
1. Create an API key
Generate a GoEnhance API key and send it in the Authorization header.
2
2. Choose materials
Select the first frame, last frame, ordered references, elements, or one reference video required by the shot.
3
3. Validate the budget
Enforce the 7-material limit, or the 4-material limit when a reference video is present, before submission.
4
4. Submit the job
POST to
https://api.goenhance.ai/api/v1/videos/generations with model: kling-o1 and persist the returned img_uuid. 5
5. Track completion
Poll
GET /api/v1/jobs/detail or provide a publicly accessible HTTPS custom_callback_url. 6
6. Handle terminal states
Process success and failure idempotently. Callback bodies match the job-detail response; non-200 callback responses are retried up to three times with a three-second timeout per attempt.
Build repeatable video workflows around explicit reference semantics.
Developer use cases for Kling O1
Character continuity
Use ordered reference images or reusable element IDs to keep approved subjects available to prompt templates.
Product motion
Animate a hosted product frame while your application manages validation, prompt templates, and approvals.
First-to-last transitions
Define opening and ending frames for short transitions without adding incompatible reference materials.
Motion reference
Use a feature reference video to guide motion or visual style while preserving one asynchronous task pipeline.
Video editing
Use base-video mode for prompt-directed edits to one supplied clip within O1’s documented duration and resolution limits.
Localized creative variants
Parameterize prompts and aspect ratios for regional campaigns while keeping task submission and completion handling consistent.
Answers to common integration, pricing, and Kuaishou API questions.