Reka Releases Rho-1: 19B Omni-Reasoning Model Understands, Generates Video and Outputs Robot Actions

Reka has released a research preview of Rho-1, a 19B omni-reasoning model trained from scratch. A single neural network understands and generates text, images and video, reasons over them, and outputs robot actions, replacing pipelines of modality-specific models. No public weights, API or pricing yet.

• 19B model unifies text, images, video and robot actions in one context window
• Two expert weight streams share attention and one KV cache in every block
• Distillation cuts denoising from 99 to 8 steps, about 1 second per 5.3 s clip
• Trained on 320 H100s for 3 months; video capped at 672×384, no public weights