# Multi-task Transformer

Janus-Pro is an innovative autoregressive framework that unifies multimodal understanding and generation capabilities. By decoupling visual encoding paths and employing a single Transformer architecture, it resolves conflicts in the roles of visual encoders between understanding and generation.

OneFormer is the first multi-task universal image segmentation framework, achieving semantic segmentation, instance segmentation, and panoptic segmentation tasks with a single model

ruT5-large is a Russian text-to-text generation model developed by the SberDevices team, based on the T5 architecture with 737 million parameters.