RELEASE · MODELS · #1566
Reka AI releases Rho-1, a 19B omni-model for text, images, video and robot control
Reka AI published a research preview of Rho-1, a 19-billion-parameter omni-model that processes and generates text, images, continuous video, and robot control actions inside a single neural network using a shared token context without external tool calls. The model was trained on about 320 H100 GPUs over roughly three months and uses an inverse dynamics approach to extract control signals from ordinary internet videos so the same weights can drive both camera prediction and robot movements.
KEY POINTS
- Reka AI published a research preview of Rho-1, a 19-billion-parameter omni-model that processes and generates text, images, continuous video, and robot control actions inside a single neural network using a shared token context without external tool calls.
- The model was trained on about 320 H100 GPUs over roughly three months and uses an inverse dynamics approach to extract control signals from ordinary internet videos so the same weights can drive both camera prediction and robot movements.
- A single weights-based omni-model that natively handles visual, temporal, and robot control modalities could simplify multi-task pipelines and advance integrated “world model” research.
WHY IT MATTERS
A single weights-based omni-model that natively handles visual, temporal, and robot control modalities could simplify multi-task pipelines and advance integrated “world model” research.