Kling AI releases version 4.0 with stereo audio and a lighter Flash variant
Kling AI announced version 4.0 of its video generator, adding higher resolution, more precise controls and a faster Flash variant. Users can now produce up to 30 seconds of video in a single pass and supply up to ten keyframe images to guide scene changes. Prompt comprehension was expanded to 8,000 tokens.
Key points
- Prompt length expands to 8,000 tokens; Omni Reference handles up to 15 assets, including 10 images and 5 videos (30 s total).
- New Flash variant runs faster at lower cost, limited to 720p resolution and 20 seconds video; stereo audio adds two‑channel sound in 10+ languages.
The new “Omni Reference” system accepts up to 15 reference assets per generation – up to 10 images, 5 videos totalling 30 seconds, and seven other elements such as storyboards or depth‑map videos. A “Creative Recreation” feature lets users swap products or characters in existing videos while keeping other elements unchanged. Stereo audio now offers two‑channel sound and improved lip sync in more than ten languages, including German, English, Japanese, Korean and Spanish. The Flash variant runs at lower cost but is limited to 720p and 20‑second clips. Access is available through MCP and CLI, though API pricing has not been disclosed.
Kling AI drops version 4.0 with stereo audio, multilingual lip sync, and a lighter Flash variant
The Decoder · 19 October 2026
Loading the full article…
This text was published by The Decoder and written by Matthias Bastian. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Generative AI & Models
All →- MiniMax releases M3.1 Flash model at $0.15 per 1M input tokens · 4 src
- Meta releases Muse Spark Open, an 816B-parameter open-weight multimodal model · 1 src
- Meta releases Muse Spark Open, a 816B-parameter multimodal model with 1M-token context · 40 src
- Anthropic releases Claude Haiku 5.5 with 1M context and lower costs · 1 src
- Google unveils Gemini 4 Argon with million-token output window · 2 src
Comments
via GitHub Discussions