XNews

Muse Spark 1.2 Supports Multimodal Tasks from Code Generation to Robot Navigation

Muse Spark 1.2 Supports Multimodal Tasks from Code Generation to Robot Navigation

Muse Spark 1.2 supports a broad range of multimodal tasks, including turning visuals into working code and translating perception into physical action. It brings robust audio-visual understanding for video-heavy enterprise workflows. New evals and demos illustrate the model's visual understanding and reasoning capabilities, including a demo showing Muse Spark parsing multimodal observations and calling tools to guide a robot navigating an unstructured environment to find a rubber duck.

Source posts