fal-ai/kling-video/o3/standard/image-to-video

Run Kling o3 Image To Video API on fal

Kling O3 Standard image-to-video on fal.ai. Animate a transition between a start frame and an optional end frame, guided by text prompts. The more affordable sibling to O3 Pro, sharing the same model generation at a lower price point.

Features

• Videos up to 15 seconds
• Start + end frame support: animate the transition between two images
• Multi-prompt support for multi-shot narrative control
• Native audio generation (Chinese and English; other languages auto-translated)
• Strong subject and text consistency
• Scene cuts and ultra-high-definition storyboards

Note: This endpoint uses image_url (not start_image_url) — this differs from v3 endpoints.

Pricing

Generation Speed

In testing, O3 Standard completed a 5s generation in ~94 seconds, compared to ~318 seconds for a comparable O3 Pro job. Standard is roughly 3x faster than Pro.

Quick Start

Install

bash
npm install --save @fal-ai/client

bash
export FAL_KEY="YOUR_API_KEY"

Submit a request

javascript
import { fal } from "@fal-ai/client";

const result = await fal.subscribe("fal-ai/kling-video/o3/standard/image-to-video", {
  input: {
    image_url: "https://example.com/start-frame.png",
    end_image_url: "https://example.com/end-frame.png",
    prompt: "The character walks forward slowly, camera following from behind.",
    duration: "10",
    generate_audio: false,
  },
  logs: true,
  onQueueUpdate: (update) => {
    if (update.status === "IN_PROGRESS") {
      update.logs.map((log) => log.message).forEach(console.log);
    }
  },
});

console.log(result.data.video.url);

Python

python
import fal_client

def on_queue_update(update):
    if isinstance(update, fal_client.InProgress):
        for log in update.logs:
            print(log["message"])

result = fal_client.subscribe(
    "fal-ai/kling-video/o3/standard/image-to-video",
    arguments={
        "image_url": "https://example.com/start-frame.png",
        "end_image_url": "https://example.com/end-frame.png",
        "prompt": "The character walks forward slowly, camera following from behind.",
        "duration": "10",
        "generate_audio": False,
    },
    with_logs=True,
    on_queue_update=on_queue_update,
)
print(result)

Input Parameters

Start + End Frame

Supply both image_url and end_image_url to animate a transition between two specific frames. The model interpolates the motion between them while following your text prompt.

javascript
{
  image_url: "https://example.com/scene-open.png",
  end_image_url: "https://example.com/scene-close.png",
  prompt: "The door slowly swings open. Dust motes drift in the light.",
  duration: "5",
}

Multi-Prompt (Multi-Shot)

Divide the video into multiple shots, each with its own prompt and duration:

javascript
{
  image_url: "https://example.com/start.png",
  multi_prompt: [
    { prompt: "Wide shot: the figure approaches the gate.", duration: "5" },
    { prompt: "Close-up: her hand reaches for the handle.", duration: "5" },
  ],
  shot_type: "customize",
  duration: "10",
}

Output

json
{
  "video": {
    "url": "https://v3b.fal.media/files/...",
    "content_type": "video/mp4",
    "file_name": "output.mp4",
    "file_size": 12037975
  }
}

Infrastructure

• Concurrency alias: fal-ai/kling-video-v3 (shared across all v3/o3 endpoints)
• Default concurrency limit: 1 per user (overrides available on request)
• For long jobs, use fal.queue.submit + webhook rather than blocking

O3 Standard vs. O3 Pro

Related Endpoints