whereismyai
Kandinsky 6
Kandinsky 6 is an open-source family of diffusion models that generates synchronized video clips with 44 kHz audio from text or images.
Kandinsky 6.0 is a family of foundation diffusion models designed for synchronized text-to-audio-video and image-to-audio-video generation. It comprises Kandinsky 6.0 Video Lite with 3B parameters and Kandinsky 6.0 Video Pro with 29B parameters. Both models generate 5-second video clips complete with synchronized 44 kHz audio and lip-sync capabilities. A plug-in super-resolution model allows outputs to reach Full HD resolution of 1920 by 1080. The project is open source and integrates with tools like Diffusers, ComfyUI, vLLM-omni, SGLang, and FastVideo.
Kandinsky 6 at a glance
- Best for: Software engineer, Designer, Social media creator
- Runs on: Web, Linux, Windows, macOS, API, self-hosted, open source
- Pricing: Free
- Released: October 2026
- Category: AI video tool
- Website: github.com/kandinskylab/kandinsky-6
Key features
- Text-to-audio-video generation
- Image-to-audio-video generation
- Synchronized 44 kHz audio and lip-sync
- Plug-in super-resolution for Full HD output
- Lite model with 3 billion parameters
- Pro model with 29 billion parameters
- Integration with ComfyUI and Diffusers
Use it for
- Generate 5-second video clips from text prompts
- Generate synchronized audio and video from images
- Upscale video output to Full HD resolution
Alternatives to Kandinsky 6
- TurboDiffusion — Accelerates video diffusion generation by up to 200 times on single GPUs.
- Fliki — Generates narrated videos from text or prompts using 2,000+ AI voices.
- Bernini — Combines a semantic planner and DiT renderer for video generation and editing tasks.
- Lottie Skill — It allows AI agents to edit, animate, and manage complex video projects via code.
- Wan — Provides native 30-second video generation and pixel-perfect consistency.
- OpenMontage — Turns your AI coding assistant into a full video production studio.
- Kling — Cinematic video model behind award-winning AI ads
- MAGI-1 — Generates synchronized audio and visuals with high-fidelity camera control.
- FreeVideo — Runs MiniMax H3 locally on consumer GPUs with as little as 8 GB of VRAM.
- PixVerse — Own video models with agent-directed shots and fast generation
- Sisif — It offers a pay-per-clip pricing model with no subscriptions or watermarks.
- FastVideo — Provides end-to-end post-training and real-time inference for accelerated video generation.
All Kandinsky 6 alternatives