Skip to main content

​
Overview

The Flow Builder engine is a single-loop state machine that processes one user message at a time. Understanding it helps you build faster, more efficient flows — especially for voice where latency matters.

​
The Processing Loop

When a user sends a message:
  1. Receive message — Input is added to conversation history
  2. Process current node — Engine executes the active node
  3. Chain through silent nodes — Variable, Logic, Extraction, Function, Request nodes execute automatically in sequence
  4. Stop at Conversation or End — Loop stops when it reaches a response-generating node
Only Conversation nodes produce user-facing responses. All other nodes execute silently. The engine can chain through multiple nodes in a single turn, transparently to the user.

​
Node Processing Order

​
Streaming (SSE)

Flow Builder uses Server-Sent Events to stream responses in real-time — critical for voice where TTS starts speaking as soon as the first tokens arrive.

​
SSE Event Types

The clear event is key for voice: when the LLM starts responding but then transitions, clear tells TTS to stop the partial response and prepare for the new node’s output.

​
Safety Limits

Hitting the 10-transition limit means there’s likely a loop — check for Logic nodes that cycle without a Conversation node in between.

​
LLM Resolution Order

The engine determines which model to use by checking (in order):
  1. Node override — Model set on the specific node
  2. Flow default — Model set at the flow level
  3. Fallback — gpt-4o-mini
Supported providers: OpenAI, Anthropic, and OpenRouter.

Conversation Node

The core node type

API Key Setup

Configure your LLM provider

Voice Connection

Deploy to voice with streaming

Testing

Test your flow