One key for all versions
Call v1.0 through v2.1 — and every other video model — with the same key, the same endpoint, and one invoice.
Use the Kling API via RunAPI with a model skill, unified auth, and pay-as-you-go pricing.
curl -X POST https://runapi.ai/api/v1/kling/motion_control \
-H "Authorization: Bearer $RUNAPI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "kling-v3-turbo-text-to-video",
"video_url": "https://cdn.runapi.ai/public/samples/video.mp4",
"prompt": "Create a 10-second video of a person walking through an autumn forest, leaves falling, cinematic camera pan."
}'
import { KlingClient } from "@runapi.ai/kling";
const client = new KlingClient();
const result = await client.motionControl.run({
model: "kling-v3-turbo-text-to-video",
video_url: "https://cdn.runapi.ai/public/samples/video.mp4",
prompt: "Create a 10-second video of a person walking through an autumn forest, leaves falling, cinematic camera pan.",
});
<?php
require __DIR__ . "/vendor/autoload.php";
use RunApi\Kling\KlingClient;
$client = new KlingClient();
$result = $client->motionControl->run([
'model' => 'kling-v3-turbo-text-to-video',
'video_url' => 'https://cdn.runapi.ai/public/samples/video.mp4',
'prompt' => 'Create a 10-second video of a person walking through an autumn forest, leaves falling, cinematic camera pan.',
]);
require "runapi/kling"
client = RunApi::Kling::Client.new
result = client.motion_control.run(
model: "kling-v3-turbo-text-to-video",
video_url: "https://cdn.runapi.ai/public/samples/video.mp4",
prompt: "Create a 10-second video of a person walking through an autumn forest, leaves falling, cinematic camera pan."
)
npx skills add runapi-ai/kling -g
# Claude Code
claude mcp add runapi -s user -- npx -y @runapi.ai/mcp
# Codex
codex plugin install runapi-mcp@agents
# Cursor / Windsurf / VS Code
npx @runapi.ai/mcp init cursor
Kling is Kuaishou's flagship video generation family, built for shot-level control: camera moves, subject motion, and consistent physics across a clip. It handles both pure text prompts and image-driven animation, which makes it a common choice for product shots, short-form ads, and previsualization.
Each version stays available after a new one ships, so you can pin a version for reproducibility or move up when quality matters more than cost. v2.1 leads on motion coherence and prompt adherence; v1.6 and below remain useful for high-volume drafts.
| Version | Billing | Pricing | |
|---|---|---|---|
| kling-3.0 | second | $0.430 / second | View → |
| kling-ai-avatar-pro | second | $0.160 / second | View → |
| kling-ai-avatar-standard | second | $0.080 / second | View → |
| kling-ai-avatar-v1-pro | second | $0.160 / second | View → |
| kling-o1 | second | $0.080 / second | View → |
| kling-v1-avatar-standard | second | $0.080 / second | View → |
| kling-v2.1-master-image-to-video | second | $0.610 / second | View → |
| kling-v2.1-master-text-to-video | second | $0.610 / second | View → |
| kling-v2.1-pro | second | $0.100 / second | View → |
| kling-v2.1-standard | second | $0.050 / second | View → |
| kling-v2.5-turbo-image-to-video-pro | call | $0.160 / call | View → |
| kling-v2.5-turbo-text-to-video-pro | second | $0.160 / second | View → |
| kling-v2.6 | second | $0.310 / second | View → |
| kling-v3-omni | second | $0.910 / second | View → |
| kling-v3-omni-edit | second | $0.910 / second | View → |
| kling-v3-omni-reference | second | $0.910 / second | View → |
| kling-v3-turbo-image-to-video | second | $0.180 / second | View → |
| kling-v3-turbo-text-to-video | second | $0.180 / second | View → |
| Endpoint | Resolution | Duration | Price |
|---|---|---|---|
| text_to_video | 720p, 1080p | 3s, 4s, 5s, 6s, 7s, 8s, 9s, 10s, 11s, 12s, 13s, 14s, 15s | $0.18-$0.22 / second |
Prices shown for kling-v3-turbo-text-to-video. Earlier versions bill at their own per-second rate.
Sign up and generate an API key from the dashboard.
Choose kling-v2.1 for best quality or an earlier version for lower cost.
POST to /api/v1/kling/text_to_video with your prompt, duration, and aspect ratio.
Poll the task or set a webhook. Download the MP4 when the task completes.
Kuaishou has shipped video models on a fast cadence since 2024, and Kling sits alongside Veo, Seedance, and Runway as one of the few families delivering reliable 1080p motion at production scale. On RunAPI it runs behind the same task API as every other video model, so switching families is a one-line model change.
Call v1.0 through v2.1 — and every other video model — with the same key, the same endpoint, and one invoice.
Aggregated volume, passed through. No subscription, no minimum spend, no per-seat fee.
Requests reroute on provider errors, and completed tasks can be pushed to your webhook instead of polled.
Text-to-video and image-to-video clips up to 10 seconds, at 720p or 1080p, with camera and motion control.
Typically 60–120 seconds for a 5-second clip; 10-second clips take about 2–3 minutes.
v2.1 has the strongest motion coherence and prompt adherence. Earlier versions are cheaper and faster, and remain available indefinitely.
Yes. Use the image_to_video task type with an image_url and an optional motion prompt.
Per second of output video. A 5-second 1080p clip on kling-v2.1 costs $2.15. You are only charged on successful completion.
Outputs are stored for 7 days so you can download them, then deleted. Prompts are not used for training.
We're here to help with enterprise setup, integrations, and technical questions.