Skip to content
Tigy AI
/
Legal centerDashboard
OverviewDeveloper
Overview

Get started

  • Introduction
  • Overview
  • First agent
  • Tools and data
  • Concepts
    • How it works
    • Calls and runs

Configure

  • Behavior
    • Prompting
    • Conversation flow
  • Voice and models
    • LLM
    • Voice
    • Transcription
    • Fallbacks
    • Audio library
  • Tools
    • Tools
    • HTTP API
    • MCP
    • Credentials
    • Call transfer
    • End call
  • Knowledge base
  • Testing
  • Agent ID

Deploy

  • Publishing
  • Versions
  • Telephony
    • Introduction
    • Connect telephony
    • Inbound calls
    • Twilio
    • Telnyx
    • SIP

Monitor

  • Runs and usage
  • Reports
  • Troubleshooting

Workspace

  • Workspace
  • Members and permissions
  • Billing
  • Security
  • Privacy

Developer

  • Call data
Dashboard
Explore pages
Overview

Get started

  • Introduction
  • Overview
  • First agent
  • Tools and data
  • Concepts
    • How it works
    • Calls and runs

Configure

  • Behavior
    • Prompting
    • Conversation flow
  • Voice and models
    • LLM
    • Voice
    • Transcription
    • Fallbacks
    • Audio library
  • Tools
    • Tools
    • HTTP API
    • MCP
    • Credentials
    • Call transfer
    • End call
  • Knowledge base
  • Testing
  • Agent ID

Deploy

  • Publishing
  • Versions
  • Telephony
    • Introduction
    • Connect telephony
    • Inbound calls
    • Twilio
    • Telnyx
    • SIP

Monitor

  • Runs and usage
  • Reports
  • Troubleshooting

Workspace

  • Workspace
  • Members and permissions
  • Billing
  • Security
  • Privacy

Developer

  • Call data
Dashboard

Documentation / LLM

LLM

Voice Agents use LLM (Large Language Models), which are trained to understand the conversational context, and respond to users.

Tigy AI provides a managed LLM for every voice agent. The model and provider are selected and maintained by Tigy AI; users do not choose or configure an external LLM.

Warning

LLM providers receive the data needed to generate responses, such as prompts, conversation history, model settings, and any configured tool/function definitions or tool call context. Review the provider's data processing, retention, model training, and regional hosting policies before using sensitive data.

← Back to documentation
© 2026 Tigy AItigyai.com ↗

On this page