Kling AI released version 4.0 on October 19, 2026, saying it improves video generation with higher resolution and more precise controls. It is the company's first major update since its previous release, which introduced foundational features for AI video creation.
Kling AI reported a prompt comprehension expansion to 8,000 tokens, measured on the company's internal benchmarks. That compares with the earlier limit of 2,000 tokens in version 3.0.
Version 4.0 is built on the company's proprietary AI architecture and targets professional video creation. Availability begins with a public beta, initially for creators and developers.
"The upgraded 'Omni Reference' system accepts up to 15 reference assets in a single generation," said Matthias Bastian, the company's lead product manager. The system includes up to 10 images, 5 videos with a combined maximum of 30 seconds, and 7 elements.
The announcement follows a competitive release from Chinese rivals, including Wan 3.0 and Seedance 2.5. Kling AI said its update holds its own but doesn't pull ahead, especially in logical shot sequencing, physics, and natural-sounding dialogue.
Kling AI did not say when pricing for the API will be announced, and the company raised questions about the next releases from its competitors. The company said it will provide more details in the coming weeks.
Source: thedecoder
