Kling 2.0

What Kling 2.0 is

Kling 2.0 is an AI video generation model from Kuaishou, the Chinese short-video company. It was launched globally in April 2025. The South China Morning Post reported on 15 April 2025 that Kuaishou had unveiled what it billed as the "world's most powerful" AI video generator, presented as a rival to OpenAI's Sora^[1][2]. The framing matters: the release was pitched not as a narrow feature update but as a claim to the front of the field in turning text into video.

Where it came from

Kling 2.0 arrived after a year in which Kling AI had already become a commercial service rather than a demo. Kuaishou said the annualized revenue run rate of Kling AI passed USD 100 million in March 2025, in the tenth month since the product launched^[3]. In the first quarter of 2025, Kling generated more than RMB 150 million of revenue^[1]. That base is the context for the 2.0 release: it was an upgrade to something that already had paying users and a business attached to it.

What changed

Kuaishou described Kling 2.0 as a significant upgrade and said it held a leading position in three areas: motion quality, semantic responsiveness and visual aesthetics^[1]. Semantic responsiveness means how faithfully a generated clip follows what the prompt actually asked for; motion quality covers how convincing movement looks once the model starts generating frames. The company repeated that leadership claim when it described the global rollout of the model^[3].

The release also introduced a concept Kuaishou calls the Multi-modal Visual Language (MVL)^[1]. Alongside it came a multimodal editing feature: users can combine images, video, voice and motion paths, and can add, remove or replace elements inside a video that has already been generated^[1]. That is a practical shift. Earlier generations of video models worked mostly one shot at a time — you write a prompt, you get a clip, and if one thing is wrong you start over. Editing inside a generated video moves the model closer to a production tool rather than a slot machine.

Access at launch was narrow. QbitAI reported that Kling 2.0 initially offered only a Master edition, priced at RMB 10 for a five-second 1080p clip^[4].

Reception and use

Kuaishou's CEO said Kling AI 2.0 was launched in April 2025 and met a strong reception globally^[1]. By May 2025 the company said the model was being widely applied in advertising, short plays and smart devices^[1] — three areas where short video is already the working format, which makes them natural first markets for a video generator. Kuaishou's management said the 2.0 launch reinforced its conviction that Kling AI could become foundational infrastructure: a base layer that other products and services build on, rather than a single app^[1].

Where it sits

Kling 2.0 was a step in a fast sequence rather than a resting point. At the end of May 2025 it was succeeded by the Kling 2.1 series^[3]. Read together with the revenue figures, the release shows how quickly Chinese video-generation models moved from research demonstrations to billed products with per-clip pricing, editing tools and named customers in advertising and drama production.