Kling: Kling v1 Text to Video API

Kling-v1/text-to-video is the pioneer foundation model of Kuaishou's video generation suite, utilizing a Large-scale Diffusion Transformer (DiT) architecture. It is the first commercially available AI to successfully generate cohesive videos up to 2 minutes long with professional-grade 1080P resolution and a standard 30fps frame rate. The model is renowned for its "High-Speed Motion Understanding" and its ability to render complex, large-scale movements (e.g., a person eating noodles, high-speed racing) with significant temporal consistency. In the 2024-2025 AI landscape, it set the benchmark for cinematic-style video generation, offering a unique "Master Mode" for enhanced artistic texture and prompt adherence.

• Input: text
• Output: video
• Released: 2025-06-05
• Knowledge cutoff: 2024-05-31