Magic Hour

« Generate 30-second native videos in a single pass, using up to 50 multimodal references (images, audio, videos) to ensure consistency in characters and scenes. Local edits are also more precise and don’t require regenerating the entire scene »

« An open-source video generator that combines text, images, video, and audio to produce 2K clips up to 15 seconds long, with native stereo sound generated in a single pass. API available »

« Generate 1080p 30fps videos up to 15 seconds long from text or images with consistent characters, cinematic camera movements, and realistic 3D physics. Features : multi-Shots, audio lip-sync,native audio, native-level text rendering, etc. »

« Generates images from text and transforms them into video »

« An AI video generator capable of producing native 4K clips up to 30 seconds long, featuring multiple linked shots, synchronized audio, and the same character appearing in each video. This AI model was created by Alibaba »

« An open-source video model that creates coherent multi-shot scenes with synchronized audio, edits real footage, and supports 4K HDR or RAW output. This video generator runs locally and is customizable»

« A single AI model to create realistic images, videos up to 20 seconds long, and native audio. Video generation includes native multilingual audio: dialogue, sound effects, and ambient sound are produced directly alongside the images, without the need for separate post-processing »

« Generate, edit, and extend videos from simple text or multimodal prompts using Google’s most advanced video model, integrated into the Gemini suite »