Kyutai's REKI model is a 2.4B vision-language-action model that outperforms larger models on benchmarks due to its efficient memory usage.

Kyutai has released REKI, a 2.4B vision-language-action model for robot manipulation. The model uses efficient memory to remember recent frames and beats larger models on three out-of-distribution benchmarks. All weights are available on Hugging Face. https://www.marktechpost.com/2026/10/17/meet-kyutai-reki-the-2-4b-vla-model-bringing-efficient-memory-to-robot-manipulation/ #AIagent #AI #GenAI #AgenticAI

Kyutai has released REKI, a compact 2.4B vision-language-action model for robot manipulation. Pretrained on public data, it outperforms models twice its size on most benchmarks. All weights are live on Hugging Face. https://www.marktechpost.com/2026/10/17/meet-kyutai-reki-the-2-4b-vla-model-bringing-efficient-memory-to-robot-manipulation/ #AIagent #AI #GenAI #Robotics