Kyutai Advances AI for Robot Vision and Action
The series follows Kyutai's development of AI models that combine vision, language, and action capabilities for robotics, beginning with the release of CADRE video autoencoders and culminating in the REKI 2.4B vision-language-action model. With REKI now available, Kyutai has achieved a unified model that can perceive, understand, and manipulate objects, marking a significant step toward real-time robot autonomy.
-
Kyutai releases CADRE open video autoencoders to reduce world model lag
Kyutai has released CADRE, a family of seven open causal diffusion representation autoencoders designed to reduce action latency in interactive world models. The autoencoders feature approximately…
1 source -
Kyutai releases REKI 2.4B vision-language-action model for robot manipulation
Kyutai, together with Inria and École des Ponts, announced REKI, a 2.4 B parameter vision‑language‑action (VLA) model built on a 2.2 B Qwen3.5‑2B backbone plus a 0.2 B flow‑matching action expert.…
1 source