Changelog

Kling 3.0 is a major release bringing the Omni model for character and audio consistency, 15-second cinematic video generation, and 4K image output.

Highlights

• Omni Model — Maintain the same face, voice, and style across multiple video clips via the Subject Library
• 15-Second Videos — Generate longer cinematic videos with smart shot scheduling and realistic physics
• 4K Image Quality — Stills that rival traditional photography with crisp typography and realistic textures
• Multilingual Lip-Sync — Perfect mouth-to-audio synchronization in English, Chinese, Japanese, Korean, and Spanish
• Reinforcement Learning — RL-based training for realistic physical simulation including flowing water, fabric, and anatomy

Video Generation

• Text-to-Video with aspect ratio selection (16:9, 9:16, 1:1)
• Image-to-Video with prompt guidance
• Support for 5s and 10s video durations
• Native 1080p output with audio sync
• Motion Control for face-driven video animation

Platform

• Credit-based pricing with monthly, yearly, and pay-as-you-go plans
• Stripe payment integration
• Multi-language support (English & Chinese)