Kling

#2

kuaishou · seit 2024-06-06 · 43× · zuletzt 01. Juli 2026

95

Momentum

Kling AI is a text-to-video generation model by Kuaishou Technology (Beijing), publicly available since June 2024. The current Kling 3.0 model (February 2026) is based on a proprietary Multi-modal Visual Language (MVL) architecture with a Diffusion Transformer (DiT) and generates videos at up to native 4K resolution, 60 fps, and 15 seconds duration from text or image inputs. The product offers a credit-based freemium model with web, iOS, and Android access as well as an API; over 60 million creators have generated more than 600 million videos to date. A "Custom Models" feature allows users to train custom character models by uploading 10–30 training video clips.

Preuves (43)