In simple English
This tool creates short videos with matching sound and voices from text or pictures. You type what you want to see and hear, and the program builds the clip. It includes lip movements so characters appear to speak the words you write. You need a powerful computer with a high-end graphics card to run it. The software is free to use under an open source licence.
What it is
Kandinsky 6.0 is an artificial intelligence model family that generates synchronized video and audio from text prompts or images. It includes a Lite version with three billion parameters and a Pro version with twenty-nine billion parameters.
What it is good for
Solo builders and content creators who want to generate short animated clips with matching audio and spoken dialogue directly on their own hardware without paying cloud subscription fees.
- Generates both video and synchronized audio at the same time.
- Includes lip-sync capabilities so characters speak your text.
- Provides multiple deployment paths including ComfyUI and Diffusers.
- Needs an expensive high-end graphics card with plenty of memory.
- Generation takes significant time even on powerful hardware.
- Large model weights require substantial disk space to download.
Cloud video generation subscriptions and manual video editing workflows
ComfyUI
- Comfy-Org/ComfyUI142k stars
A node-based graphical interface for running diffusion models like Kandinsky.
- huggingface/diffusers35k stars
A Python library for running pretrained image, audio, and video diffusion models.
- deepset-ai/haystack27k stars
A Python framework for joining language models, document search and tools into one working pipeline.
- Giskard-AI/giskard-oss5.9k stars
A Python library that writes tests for language-model programs and tries to break them on purpose.
- ggml-org/llama.cpp132k stars
A C and C++ library and command-line tool for running language models locally.
git clone https://github.com/kandinskylab/kandinsky-6.git cd kandinsky-6 just setup just download pro-distill just generate 'A cat is lying on a mat, turns toward the camera, opens its mouth and says <S>Hello<E>'