Mistral unveils Robostral Navigate for embodied robot navigation

The 8B vision-language-action model achieves state-of-the-art on R2R-CE, controlling robots with plain-language instructions without LiDAR or depth sensors. It was trained entirely in simulation.