POST https://fal.run/fal-ai/kling-video/o1/image-to-video
Endpoint ID: fal-ai/kling-video/o1/image-to-video
Try it in the Playground
Run this model interactively with your own prompts.
Quick Start
Dual-Keyframe Control Architecture
Kling O1’s dual-keyframe approach inverts the standard image-to-video workflow by accepting both start and end images as constraints, then synthesizing the motion path between them. Where traditional image-to-video models extrapolate freely from a single frame, this variant interpolates along a defined trajectory while respecting prompt-driven style guidance. What this means for you:- Deterministic endpoints: Generate 5 or 10-second videos where both opening and closing frames match your exact specifications, critical for brand-consistent sequences or narrative continuity
- Controlled transformation arcs: Animate seasonal changes, product evolutions, or character state transitions with predictable motion paths rather than AI-improvised endings
- Prompt-guided interpolation: Reference start and end frames via
@Image1and@Image2in prompts to direct style, camera movement, and scene dynamics during the transition - Flexible duration options: Choose between 5-second (1.12) outputs to match pacing requirements for social media clips or longer editorial sequences
Performance Scaling
Kling O1 operates within Kuaishou’s broader Kling Video ecosystem, prioritizing frame-accurate transitions over generation speed.| Metric | Result | Context |
|---|---|---|
| Duration Options | 5s or 10s | Configurable via duration parameter |
| Cost per Video | 1.12 (10s) | Based on $0.112 per second rate |
| Input Requirements | Start + end frame (both required) | Min 300px dimensions, 0.40-2.50 aspect ratio, 10MB max per image |
| Output Format | MP4 video | Single video file with synthesized motion path |
Technical Specifications
| Spec | Details |
|---|---|
| Architecture | Kling O1 |
| Input Formats | Start image (required), end image (required), text prompt with frame references |
| Output Formats | MP4 video |
| Duration Options | 5 seconds or 10 seconds |
| Prompt Syntax | Use @Image1 and @Image2 to reference uploaded keyframes in prompts |
| License | Commercial use via fal partnership |
How It Stacks Up
Kling Video Image to Video (v2.5-turbo) - Kling O1 specializes in dual-keyframe interpolation for transformation sequences where endpoint precision matters, making it ideal for timelapse effects and morphing animations. Kling Video v2.5-turbo emphasizes single-frame extrapolation with faster generation speeds for open-ended creative exploration. Kling 2.0 Master Image to Video - Kling O1 trades generation flexibility for endpoint accuracy, constraining motion paths between two defined frames at a specialized price point. Kling 2.0 Master prioritizes maximum quality single-frame animation for premium production workflows requiring cinematic motion without endpoint constraints.Related
- Kling O1 Edit Video [Pro] — Video Generation
- Kling O1 First Frame Last Frame to Video [Standard] — Video Generation
- Kling O1 Reference Image to Video [Pro] — Video Generation
- Kling O1 Reference Video to Video [Pro] — Video Generation
Capabilities
- Text prompt input
- Duration control
API Reference
Input Schema
Use @Image1 to reference the start frame, @Image2 to reference the end frame.
Image to use as the first frame of the video.Max file size: 10.0MB, Min width: 300px, Min height: 300px, Min aspect ratio: 0.40, Max aspect ratio: 2.50, Timeout: 20.0s
Image to use as the last frame of the video.
Video duration in seconds. Default value:
"5"Possible values: 3, 4, 5, 6, 7, 8, 9, 10Output Schema
The generated video.
Input Example
Output Example
Limitations
durationrestricted to:3,4,5,6,7,8,9,10