详细内容
Voice AI agents for developers
Talk to vapigive it a Try
Teamsattrail-blazingstartupstoFortune500companiesdeployvoiceagentsontheVapiplatform.
ThemostconfigurableAPItobuildleadingvoiceAIproductsandscalephoneoperations.
Calls
132M+150M+
Assistants Launched
1.3M+1.5M+
Developers
308K+350K+
API
Making voice AI simple and accessible.
TypeScriptPythoncURLReact (web SDK)
1# npm install @vapi-ai/server-sdk
2import { VapiClient } from '@vapi-ai/server-sdk';
3
4const vapi = new VapiClient({
5 token: 'YOUR_PRIVATE_API_KEY' // Get your private api key from the dashboard
6});
7
8async function createCall() {
9 const call = await vapi.calls.create({
10 phoneNumberId: 'YOUR_PHONE_NUMBER_ID', // Create a free phone number in the dashboard
11 customer: { number: '+1234567890' }, // Your customer's phone number
12 assistant: {
13 model: {
14 provider: 'openai',
15 model: 'gpt-4o',
16 messages: [\
17 {\
18 role: 'system',\
19 content: 'You are a helpful AI assistant. Keep your responses concise and friendly.'\
20 }\
21 ]
22 }
23 }
24 });
25}
26
27
Copy
Use cases
Feels human. Moves the needle.
Inbound calls
Outbound calls
Inbound calls
powers 400,000+ daily calls.
FleetWorks
saves 100s
of engineering hours monthly.
A key technical requirement was the ability to bring our own stack — Vapi's developer-friendly API-first approach made this possible.
Quang Tran
CTO, FleetWorks
Receive call
Shipper Info
Use tool
Fetch_available_jobs_from_broker
Condition
Job Available?
Integrations
Integrate with more than 40+ apps in a snap.
How It Works
Try in minutes. Deploy in days.
- 001
Choose your workflow.
Pick from 1000s of pre-made templates, or build one yourself.
- 002
Plug it in.
Whether it's in your telephony, website, or in your app, we have it covered.
- 003
Done.
Handle millions of calls and watch how they perform.
Features
Flexible for engineers. Easy for business users.
Multilingual
Talk to agents in English, Spanish, Mandarin, or one of 100+ other supported languages.
API-native
Everything is exposed as an API, with 1000s of configurations and integrations.
Automated testing
Design test suites of simulated voice agents to identify hallucination risks before going to production.
Bring your own models
Bring your own API keys for transcription, LLM, or text-to-speech models. Or, plug in your own self-hosted models.
Tool calling
Plug in your APIs as tools to intelligently fetch data and perform actions on your server.
A/B experiments
Test different variations of prompts, voices, and flows to continuously optimize performance.
ENTERPRISE
Reliable.
Scalable.
Secure.
99.99%uptime
Our custom real-time audio infrastructure operates with enterprise-grade reliability.
Forward-deployedteam
Get deployment assistance and a dedicated forward-deployed engineer to go live in a week.
Sub-500mslatency
Scale up and down to millions of calls in minutes with ultra-low latency interactions.
AIguardrails
Built-in conversation guardrails prevent model hallucinations and ensure data integrity.
SOC2, HIPAA,PCI compliant
Enterprise-level security for even the most regulated healthcare and financial services.
Community
112,924+250,000+devs are already here. Join the movement.
- CONFIGURATION POINTS\
\
4.2K+\
\
Docs - SUPPORT TOPICS\
\
13K+\
\
Community - FOLLOWERS\
\
9.6K+\
\
LinkedIn
Careers at vapi
Brilliant minds. Bleeding- edge tech. Relentless drive.
Join a team of the most cracked people to do rewarding work in an intensely ambitious environment.
Your questions, answered.
What is Vapi?
How is this more cost-effective for my organisation?
What is the difference from other AI voice competitors?
I need holistic customization, what types of support does your platform offer?
Is it difficult to set up?