Skip to main content
A modality is a channel your agent uses to perceive or respond. Zero Runtime agents are voice-first, but the same pipeline can take text in, see video, and render a visual avatar, often at the same time. Pick the modalities your experience needs; the pipeline wires them in.

Speech & Audio

Voice in, voice out, plus background audio and TTS caching.

Text

Text input and output for chatbots, omnichannel, and debugging.

Vision

Let the agent see: send camera or screen frames to the model.

Avatars

Give the agent a real-time visual face.

​
How modalities map to the pipeline

Modalities are selected by the components you pass to the Pipeline and by per-session flags on zrt.serve(...) or zrt.Room(...): Modalities compose: a pipeline can run voice, vision, and an avatar together.

​
References

​
Examples

Multimodal Agent

Combine speech, text and vision in one agent.

Cascade Basic

Voice agent built from a cascade pipeline.

Realtime Basic

Speech-to-speech realtime agent.

​
SDK Reference

Pipeline

Pipeline in the Python API reference.