# Build custom voice agents **that scale**

## Built for developers

API-native and compatible with any stack for total control.

## Configurable by design

1000+ points of config, bring your own models, or choose from 200+.

FEATURES

## Built for developers, ready for business

Build voice agents with the flexibility and customizability of raw code, without having to build the entire voice stack from scratch.

### Ultra low latency

Responses in under 500ms for seamless, real-time conversation.

### Natural conversation ability

Turn-taking, human-like voices, and custom pronunciation and transcription keywords.

### Live call tool & data access

Ground responses in external data and trigger tools, including MCP, in the moment.

### Custom model support

Bring your own models or choose from 200+ fully-supported integrations.

### Telephony control

Custom SIP support, warm transfers, voicemail detection and DTMF support.

### 99.99% Reliability

Enterprise-grade uptime with built-in recovery from model or network failures.

### Works for the whole team

Devs build exactly what they want via API—product teams move fast with UI. No compromise.

HOW IT WORKS

From silence to speech in minutes

Spin up a call-ready voice assistant simply, then customize and scale however you like.

Give your agent context

Provide instructions and tools, and configure as if you were building it yourself.

Pick your models

Choose from 200+ built-in integrations or connect your own model.

Integrate your business data

Integrate your authenticated internal API services to fetch data and perform actions.

Tune and observe in production

Catch hallucinations before they go live with evals and observability stack support.