Is Kling AI Chinese? Who owns it and where it's based
Yes. Kling AI is developed by Kuaishou Technology, a Beijing-based company listed in Hong Kong. Who owns it, where its API runs, and other access.

Yes. Kling AI is a Chinese product: it is developed by Kuaishou Technology, which Kling describes as a Beijing-based Chinese technology company listed on the Hong Kong stock exchange (HKEX: 1024). Kling gives its headquarters as Beijing, with global operations through a Singapore entity.
These facts come from Kling's own llms.txt company summary and its API authentication page, read on 2026-09-28. This post answers who makes Kling and where; it doesn't assess legal, data-transfer or safety questions.
Who owns Kling AI?
Kuaishou Technology, its parent company. Kling says it was developed by Kuaishou's AI research team and launched publicly in June 2024.
| Item | What Kling's page says |
|---|---|
| Parent company | Kuaishou Technology (HKEX: 1024) |
| Headquarters | Beijing, China (with global operations via Singapore entity) |
| Initial release | 2024-06-06 |
| Latest major version | Kling 3.0 series (2026-02-05) |
| Platform type | AI video and image generation SaaS with REST API access |
| Revenue mix | About 70% from overseas markets, as of early 2026 |
What does Kling AI make?
Kling's summary says the platform specializes in text-to-video, image-to-video and text-to-image generation, and lists its major versions (llms.txt):
- Kling 1.0 in June 2024, then 1.5 and 1.6 in late 2024.
- Kling 2.0 in April 2025, 2.1 in May 2025, and 2.5 Turbo in September 2025.
- Video O1 and Kling 2.6, which added native audio, in December 2025.
- Kling 3.0 and 3.0 Omni in February 2026.
Where does Kling's API run?
Kling lists two API endpoints: a global one at api-singapore.klingai.com and a China one at api-beijing.klingai.com (llms.txt). Its authentication page says the endpoint changed from api.klingai.com to https://api-singapore.klingai.com, which "is suitable for users whose servers are located outside of China" (Authentication). Calls use an API key sent as Authorization: Bearer.
So a developer in the US or Europe calls Kling's Singapore endpoint, not the Beijing one. Kling lists its web app at kling.ai/app for work that doesn't need the API.
Is Kling AI open source?
Kling's own summary describes it as a SaaS platform with REST API access, and its pages read for this post don't describe downloadable model weights. You use Kling through its web app or its API, or through a multi-model API that offers a Kling model.
Can I use Kling models through another API?
Yes. Sume, for example, offers Kling 3.0 next to other video models: it is the catalog id kling-3 on POST /v1/videos, billed per second from the workspace's USD balance at the provider's list price × 1.25 (Video generation): $0.14 per second without audio and $0.21 with generated audio in current pricing code, plus a 5.5% agent fee by default. Kling 3.0 API covers its limits and a sample call.
Using Kling through another company doesn't change who makes the model. For your own data questions, read each company's policy: Sume's privacy policy says Sume may process and store information in the United States and other countries where Sume or its service providers operate.
Sources
Related posts
More in Models
- AI video prompt JSON: what the keys actually do
A JSON prompt for AI video is still text: the model reads keys as words. Length, size and audio are request fields, not keys inside the prompt.
- Kling 3.0 prompt guide: length, shots, dialogue, negatives
A Kling 3.0 prompt can run to 3,072 characters and include negative wording. Kling's rules for shot lists, dialogue, languages and Elements.
- Kling 3.0 vs Kling 3.0 Omni: inputs, limits and price
Kling 3.0 and 3.0 Omni share 3–15 s clips up to 4K with native audio. Omni adds reference and edit videos; 3.0 adds Motion Control. Prices compared.
- Kling 3.0 vs Seedance 2.0: inputs, limits and price
On Sume, Kling 3.0 and Seedance 2.0 both make 4–15 s clips from text or frames. Seedance adds references and six ratios, and bills per token.
Written by Sume