What is LLM?

The language model that decides what the agent says.

A large language model generates the agent's replies from the conversation so far and the instructions it was given. On a call it sits between speech recognition and speech synthesis, and its speed contributes directly to how long the caller waits.

This is an industry glossary. A term defined here is not a statement that Drop AI provides it — see the pricing page for what is included.

System prompt

The written instructions that define an agent's behaviour.

Context window

How much the model can consider at once.

Latency

The delay before the agent starts replying.

Function calling

Letting the model trigger real actions.

Voice AI

More in this category

Voice agent Conversational AI ASR STT TTS System prompt Function calling Tool use Knowledge base RAG Grounding Hallucination

Transform every customer conversation.

Start free, or talk to us about what a custom rollout looks like for your team.

Try for Free Contact Sales