Kling API Python SDK for RunAPI

The Kling Python SDK is the language-specific package for Kling on RunAPI. Use this package for video generation, animation, and video editing workflows when your application needs request bodies, task status lookup, and consistent RunAPI errors in Python.

This README is the Python package guide inside the public kling-sdk repository. For the repository overview, start at ../README.md; for model details, use https://runapi.ai/models/kling; for API reference, use https://runapi.ai/docs/api/kling/text-to-video; for SDK docs, use https://runapi.ai/docs/resources/sdks.

Install

pip install runapi-kling

Quick start

from runapi.kling import KlingClient

client = KlingClient()  # reads RUNAPI_API_KEY, or pass api_key="sk-..."

task = client.text_to_video.create(
    model="kling-3.0",
    prompt="A cat walking through a garden",
)
status = client.text_to_video.get(task.id)

avatar = client.ai_avatar.create(
    model="kling-ai-avatar-pro",
    prompt="A friendly host greeting the audience",
    source_image_url="https://cdn.runapi.ai/public/samples/portrait.jpg",
    source_audio_url="https://cdn.runapi.ai/public/samples/voice.mp3",
)

Kling O1 reference media

result = client.text_to_video.run(
    model="kling-o1",
    prompt="Keep <<<image_1>>> beside the performer from <<<video_1>>>",
    reference_image_urls=["https://cdn.runapi.ai/public/samples/portrait.jpg"],
    reference_video_url="https://cdn.runapi.ai/public/samples/video.mp4",
    reference_video_type="feature",
    preserve_reference_video_audio=True,
    mode="pro",
    duration_seconds=5,
)

Number reference images in prompt order as <<<image_1>>>, <<<image_2>>>, and so on; the optional video is <<<video_1>>>. With a video, send at most four images. Do not combine last_frame_image_url with reference images or a reference video. A feature reference video may be used with the required first frame; base cannot be combined with frame inputs. O1 requests are five seconds and keep sound disabled. Pricing and limits: https://runapi.ai/models/kling/o1.

Kling V3 Omni source-video editing

Use kling-v3-omni-reference with text_to_video for image-only reference inputs. For a source video, use edit_video with either kling-v3-omni-reference or kling-v3-omni-edit. The reference model keeps sound disabled for source-video requests.

Use create when you want to submit a task and return quickly, get when you need the latest task state, and run when a script should create and poll until completion:

result = client.text_to_video.run(
    model="kling-3.0",
    prompt="A serene drone shot over a misty forest",
)
print(result.videos[0].url)

In web request handlers, prefer create plus webhook or later get polling so a worker is not held open.

RunAPI-generated file URLs are temporary. Download and store generated images, videos, audio, or other files in your own durable storage within 7 days; do not treat returned URLs as long-term assets.

Language notes

Pass parameters as keyword arguments and catch the runapi.kling error classes when building video jobs, workers, or scripts. The available resources are text_to_video, edit_video, ai_avatar, image_to_video, and motion_control. Keep RUNAPI_API_KEY in the environment or your secret manager; never commit API keys or callback secrets.

Links

• Model page: https://runapi.ai/models/kling
• SDK docs: https://runapi.ai/docs/resources/sdks
• Product docs: https://runapi.ai/docs/api/kling/text-to-video
• Pricing and rate limits: https://runapi.ai/models/kling/3.0
• Provider comparison: https://runapi.ai/providers/kuaishou
• Full catalog: https://runapi.ai/models
• Repository: https://github.com/runapi-ai/kling-sdk

License

Licensed under the Apache License, Version 2.0.