Embedding Overview

The same pattern everywhere

In every host, three things stay the same:

Construct a DialogMachine with a flow, an LLM URI, and tools
Route inbound text to dialog_machine.turn(text)
Send the reply text back to the host’s output channel

The host varies; the SuperDialog code is identical.

# This object works in every host below
from superdialog import DialogMachine, Flow

dialog_machine = DialogMachine(
    flow=Flow.load("kyc.json"),
    llm="anthropic/claude-haiku-4-5",
    tools=[...],
)

Choose your host

CLI chatbot

Zero infrastructure. Best for testing and prompt tuning.

LiveKit

Voice agent via Agent(llm=DialogMachineLLM(...)) plugin.

PipeCat

Drop-in FrameProcessor for PipeCat pipelines.

FastAPI

REST endpoint for text chatbots and web widgets.

Unpod Voice

Plug your DialogMachine into an Unpod AgentRunner session — no extra server needed.

Unit testing

Because SuperDialog is text-only, every dialog is unit-testable.

Lines of code comparison

Host	Adapter	Extra LoC
CLI	None - direct `input()`/`print()` or `superdialog chat`	~5
LiveKit	`DialogMachineLLM`	~8
PipeCat	`make_processor`	~12
FastAPI	`FastAPIRouter` or direct route	~6
Unpod Voice (SDK)	`unpod.AgentRunner` + `session.dialog_machine`	~6
Unit test	None - direct calls	~3
Custom (Slack, Discord, etc.)	None - direct callback	~3

CLI Reference Unpod Voice Infrastructure

​The same pattern everywhere

​Choose your host