Kling 模型

瀏覽 Kling 的模型

Kling AI is a series of advanced generative models supporting speech synthesis and digital humans: - **Kling Image**: The image generation large model that best understands Chinese, trained on LLM large language models and billions of Chinese corpora, with comprehensive metrics surpassing open-source models like SDXL/SD3 and closed-source models like Midjourney, achieving SOTA (state-of-the-art) effects in Chinese scenarios, supporting painting generation. Kling Video: The world's leading video generation large model, Kling is the first video generation large model with effects comparable to Sora and open to users, holding a leading position both domestically and internationally. Supports multiple versions: V1 (std-5s; std-10s; pro-5s; pro-10s); V1.5/V1.6 (std-5s; std-10s; pro-5s; pro-10s); v2-1; v2-master/v2.1-master; v2-5-turbo (pro only supported) - **Kling Extend**: Video extension function, supports V1 versions (std: 1x; pro: 3.5x). - **Kling Lip Sync**: Lip-sync video generation, supports 5 seconds (1x) and 10 seconds (2x) duration. - **Kling Virtual Try-On**: Virtual try-on function, supports painting and clothing generation. - **Kling Effects**: Video effects generation tool.