This evergreen history article uses authoritative archives and official records. Exact dates are used when documented; gradual inventions and rollouts are described as periods rather than being assigned a misleading single birthday.
Quick facts
- Kuaishou introduced Kling publicly in China in June 2024 as a self-developed video-generation model.
- The early model attracted attention for realistic motion, physical-world simulation and relatively long video generation.
- Kling expanded from text-to-video into image-to-video, video extension, multi-image reference, motion control, editing and sound generation.
- By 2026, Kling had become a broader image, video and audio creative studio used by individual creators and professional production teams.
- Its rapid development made Kling one of the most visible Chinese competitors in the global generative-video market.
Kuaishou’s path into generative video
Kuaishou began as a Chinese short-video and social platform, which gave the company extensive experience in video recommendation, computer vision, effects, creator tools and large-scale media infrastructure. As generative AI accelerated after 2022, Kuaishou invested in foundation models that could understand and create moving images rather than only classify or recommend them.
That background shaped Kling. The system was designed not merely as a research demonstration but as a practical creation tool connected to the workflows of short-video makers, advertisers, animators and filmmakers.
The June 2024 launch
Kuaishou announced the Kling large model in June 2024. It said the system could transform text prompts into high-quality video while modelling complex movement and aspects of physical interaction. The original demonstration emphasized 1080p output, different aspect ratios, coherent motion and clips longer than many early text-to-video competitors could produce.
Initial access was limited, but the demonstrations quickly spread internationally. In July 2024 Kuaishou opened a web version and broader global beta testing, turning Kling from a China-focused model into an internationally accessible creative product.
From one model to a creative platform
Kling’s development moved quickly through frequent model and product iterations. Image-to-video allowed creators to animate still images. Video extension made it possible to continue a scene. Multi-image reference improved character, object and style consistency. Motion-control and camera tools gave users more direct influence over movement and composition.
Kling 2.0 introduced a more multimodal workflow in which text, images and video references could be combined to express a creative intention. Later releases added stronger editing, audio-visual generation and unified generation systems. By the Kling 3.0 generation in 2026, the service presented itself as an all-in-one studio for video, images, sound and multi-shot storytelling.
Why Kling became important
Kling showed that the leading AI-video race was not limited to a few American laboratories. It combined aggressive iteration with creator-friendly access and pricing, helping professional-looking AI video reach a much broader audience.
The platform also demonstrated how a company with an existing video ecosystem could connect model research to real production needs. Marketing, e-commerce, animation, games, short drama and film production became important use cases rather than distant experiments.
Limits, safety and creative debate
Like every generative-video system, Kling can produce visual errors, inconsistent details and unrealistic interactions. It also raises questions about consent, copyrighted training material, impersonation, misinformation and the effect of automation on creative work.
The long-term value of Kling will therefore depend not only on resolution and realism but also on provenance, moderation, commercial rights, controllability and the ability to keep human creators in charge of the final result.
Kling in 2026 and beyond
By 2026, Kling had moved well beyond its original launch. Kuaishou reported strong commercial growth, broader professional adoption and major investment in the business. The product combined generation, editing, references, audio and collaborative creation in one environment.
The next stage is likely to focus on longer narrative continuity, better character memory, interactive editing, production APIs and stronger integration with advertising and filmmaking pipelines.
Timeline
| Year | Location | Event | Why it mattered |
|---|---|---|---|
| June 2024 | Beijing, China | Kuaishou unveils the Kling video-generation model | Introduced a major Chinese competitor in realistic text-to-video creation. |
| July 2024 | Global web markets | Kling opens wider beta access on the web | Made the system accessible to creators outside China. |
| 2024–2025 | Global | Image-to-video, extension and reference tools expand | Turned the model into a practical creator workflow. |
| April 2025 | China and global markets | Kling 2.0 introduces multimodal visual-language controls | Allowed text, image and video references to guide generation together. |
| 2025 | Global | Motion control, unified editing and audio features grow | Improved professional control and storytelling consistency. |
| February 2026 | Global | Kling 3.0 series launches | Expanded the product into a broader cinematic image-video-audio studio. |
Frequently asked questions
Who created Kling AI?
Kling AI was created by Kuaishou Technology, the Chinese company known for its short-video and social platforms.
When was Kling AI launched?
Kuaishou publicly introduced Kling in June 2024 and expanded global web access in July 2024.
What can Kling AI generate?
Depending on the model and product version, Kling can generate and edit video, animate images, use references, control motion and camera behavior, and create audio-visual content.
Why is Kling AI important?
It became one of the strongest global examples of a creator-facing AI-video platform developed in China and helped intensify competition in generative filmmaking.