Kandinsky Lab Open Sources Video AI With 29B Parameter Pro Model

AI15 days ago

The creators of the Kandinsky 6.0 suite made available new tools to generate video with native synchronized audio and high-fidelity visuals. The release consists of two models, a 29B parameter Pro version for cinematic quality and a 3B parameter Lite version for fast iteration, with both the code and weights provided under an MIT license.

The model is currently available via the fal platform, which includes integrated upscaling and standalone video super-resolution tools. Additionally, the vLLM-Omni project provides day 0 inference support, allowing the system to generate 5 second video clips from text or image inputs with lip sync capabilities.

Mapped to your portfolio

Possible scenarios

40%Open Model Adoption+0.3%
30%Sanctions Limit Scaling+0.1%
20%Policy Backlash-0.2%

Each figure is the modeled move of the S&P 500 if the outcome occurs. For information only. Not investment advice.

Looking for more context or analysis?

Tilt chat provides reliable, tailored insights, alerts, and automations.

Kandinsky Lab Open Sources Video AI With 29B Parameter Pro Model

In this story

Kandinsky Lab Fal.ai vLLM-Omni

Portfolio

SPY
State Street SPDR S&P 500 ETF Trust
100%

Portfolio

SPY
State Street SPDR S&P 500 ETF Trust
100%

Mapped to your portfolio

Possible scenarios

40%Open Model Adoption+0.3%
30%Sanctions Limit Scaling+0.1%
20%Policy Backlash-0.2%

Each figure is the modeled move of the S&P 500 if the outcome occurs. For information only. Not investment advice.