Home
Vapi
Vapi
Vapi is a San Francisco-based voice AI platform that lets developers build real-time, low-latency voice agents over phone, web, and SIP. It orchestrates three modular components — a transcriber (STT), an LLM, and a voice (TTS) — into a sub-700ms voice-to-voice pipeline, with first-class support for tools/function calling, multi-agent squads, outbound campaigns, persistent sessions, structured outputs, recording artifacts, evaluation scorecards, and a full REST API plus official SDKs in TypeScript, Python, C#, Ruby, PHP, and client SDKs for Web, React Native, iOS, and Python. Vapi is a Y Combinator company.
15 APIs
15 Capabilities
16 Features
AI Voice AI Voice Agents Conversational AI Telephony Real-Time Transcription Text-to-Speech LLM Agents MCP
Vapi publishes 15 APIs on the APIs.io network, including Assistants API, Calls API, Phone Numbers API, and 12 more. Tagged areas include AI, Voice AI, Voice Agents, Conversational AI, and Telephony.
The Vapi catalog on APIs.io includes 15 machine-runnable capabilities , 1 JSON-LD context, and 1 Spectral governance ruleset.
Vapi’s developer surface includes developer portal, documentation, getting-started guide, signup flow, pricing, engineering blog, changelog, and 31 more developer resources.
Create and manage Vapi voice assistants. An assistant bundles a transcriber, LLM, voice (TTS), server URL, tools, first message, system prompt, voicemail handling, and call anal...
Create outbound calls, accept inbound calls, list and filter calls, fetch call status and analysis, delete call data, and download per-call artifacts including mono and stereo r...
Buy, BYO-import, and manage phone numbers from Vapi (vapi.phoneNumber), Twilio, Telnyx, Vonage, and SIP trunks. Attach an assistant or squad to a number for inbound routing and ...
Compose multiple assistants into a Squad — a multi-agent workflow where the active assistant can hand off a live call to another assistant in the squad based on tools, intents, ...
Define and manage Tools (function tools, transfer-call tools, end-call tools, DTMF tools, hangup tools, voicemail tools, MCP tools, query tools, Make/Zapier tools, knowledge-bas...
Upload, list, retrieve, update, and delete files used as inputs to knowledge bases and other assistant-side artifacts.
Text-based conversation surface that reuses the same assistant configuration as voice calls, including an OpenAI-compatible /chat/responses endpoint for drop-in replacement of O...
Stateful sessions that persist conversation history across calls and chats so an assistant can pick up where the customer left off.
Outbound calling campaigns — upload a customer list and schedule batched outbound calls against one or more assistants and phone numbers with concurrency and retry policies.
Run aggregate analytics queries across calls — group by assistant, time bucket, status, end-reason, success evaluation, cost, and duration — for dashboards and chargeback report...
Define and run Insight queries across call data, including preview-before-save and on-demand runs, to extract structured signals from conversation transcripts.
Define LLM-graded scorecards that automatically evaluate every call against your defined success criteria and produce structured scores for trend analysis and regression detection.
Create test suites that simulate end-user behavior against an assistant, run them on demand, and inspect per-run pass/fail and judge output for regression testing of voice agents.
Define JSON-Schema-driven structured outputs that the assistant must populate from a call, plus an on-demand /structured-output/run endpoint to backfill outputs against an exist...
Generic CRUD over provider-side resources (e.g. ElevenLabs voices, Cartesia voices, Deepgram custom models) so they can be created and managed through Vapi without leaving the p...
Run Capabilities with Naftiko — Deploy and orchestrate these API capabilities using Naftiko Fleet.
Run with Naftiko
Vapi Analytics API — Analytics. 1 operations. Self-contained Naftiko capability covering the Vapi Analytics business surface.
Run with Naftiko
Vapi Assistants API — Assistants. 5 operations. Self-contained Naftiko capability covering the Vapi Assistants business surface.
Run with Naftiko
Vapi Calls API — Calls. 12 operations. Self-contained Naftiko capability covering the Vapi Calls business surface.
Run with Naftiko
Vapi Campaigns API — Campaigns. 5 operations. Self-contained Naftiko capability covering the Vapi Campaigns business surface.
Run with Naftiko
Vapi Chats API — Chats. 5 operations. Self-contained Naftiko capability covering the Vapi Chats business surface.
Run with Naftiko
Vapi Eval API — Eval. 9 operations. Self-contained Naftiko capability covering the Vapi Eval business surface.
Run with Naftiko
Vapi Files API — Files. 5 operations. Self-contained Naftiko capability covering the Vapi Files business surface.
Run with Naftiko
Vapi Insight API — Insight. 7 operations. Self-contained Naftiko capability covering the Vapi Insight business surface.
Run with Naftiko
Vapi Observability/Scorecard API — Observability/Scorecard. 5 operations. Self-contained Naftiko capability covering the Vapi Observability/Scorecard business surface.
Run with Naftiko
Vapi Phone Numbers API — Phone Numbers. 6 operations. Self-contained Naftiko capability covering the Vapi Phone Numbers business surface.
Run with Naftiko
Vapi Provider Resources API — Provider Resources. 5 operations. Self-contained Naftiko capability covering the Vapi Provider Resources business surface.
Run with Naftiko
Vapi Sessions API — Sessions. 5 operations. Self-contained Naftiko capability covering the Vapi Sessions business surface.
Run with Naftiko
Vapi Squads API — Squads. 5 operations. Self-contained Naftiko capability covering the Vapi Squads business surface.
Run with Naftiko
Vapi Structured Outputs API — Structured Outputs. 6 operations. Self-contained Naftiko capability covering the Vapi Structured Outputs business surface.
Run with Naftiko
Vapi Tools API — Tools. 5 operations. Self-contained Naftiko capability covering the Vapi Tools business surface.
Run with Naftiko
Run Capabilities with Naftiko — Deploy and orchestrate these API capabilities using Naftiko Fleet.
Run with Naftiko
Sub-700ms voice-to-voice latency orchestration layer
Transcriber + LLM + voice (TTS) modular pipeline with multi-provider choice
Bring your own provider keys to pay providers at cost
Assistants, Squads (multi-agent handoff), Sessions (persistent state)
Inbound and outbound calling over Twilio, Telnyx, Vonage, vapi.phoneNumber, and SIP
Campaigns for batched outbound calling with concurrency/retry
Tools (function, transfer-call, end-call, DTMF, hangup, voicemail, MCP, query, Make, Zapier, GHL, KB)
Server URL webhooks for real-time call events and function execution
Per-call mono/stereo/customer/assistant/video recordings, PCAP packet captures, call logs
Analytics, Insights, Observability Scorecards, and Evals for production monitoring and regression testing
Structured Outputs for JSON-schema-constrained extraction from calls
Chats API (text channel) and OpenAI-compatible /chat/responses
Official SDKs - server (TS, Python, C#, Ruby, PHP) and client (Web, Python, React Native, iOS, HTML)
Vapi MCP Server for Model Context Protocol integration
HIPAA and Zero Data Retention add-ons; SOC2 and PCI on Scale plan
$0.05/min Vapi orchestration fee on Build plan; model and voice costs passed at cost
0 classes · 12 properties
JSON-LD
17 rules ·
11 errors
6 warnings
SPECTRAL
Sources
aid: vapi-ai
url: https://raw.githubusercontent.com/api-evangelist/vapi-ai/refs/heads/main/apis.yml
apis:
- aid: vapi-ai:vapi-assistants-api
name: Vapi Assistants API
tags:
- Assistants
- Voice AI
- Agents
humanURL: https://docs.vapi.ai/api-reference/assistants/list
properties:
- url: https://docs.vapi.ai/api-reference/assistants/list
type: Documentation
- url: openapi/vapi-assistants-api-openapi.yml
type: OpenAPI
- url: json-schema/vapi-assistant-schema.json
type: JSONSchema
- url: json-structure/vapi-assistant-structure.json
type: JSONStructure
- url: examples/vapi-assistant-example.json
type: Example
- type: NaftikoCapability
url: capabilities/assistants-assistants.yaml
description: Create and manage Vapi voice assistants. An assistant bundles a transcriber, LLM, voice (TTS),
server URL, tools, first message, system prompt, voicemail handling, and call analysis plan into a reusable
voice agent configuration that can be attached to inbound or outbound phone numbers, web calls, or squads.
- aid: vapi-ai:vapi-calls-api
name: Vapi Calls API
tags:
- Calls
- Voice AI
- Telephony
humanURL: https://docs.vapi.ai/api-reference/calls/list
properties:
- url: https://docs.vapi.ai/api-reference/calls/list
type: Documentation
- url: openapi/vapi-calls-api-openapi.yml
type: OpenAPI
- url: json-schema/vapi-call-schema.json
type: JSONSchema
- url: json-structure/vapi-call-structure.json
type: JSONStructure
- url: examples/vapi-call-example.json
type: Example
- type: NaftikoCapability
url: capabilities/calls-calls.yaml
description: Create outbound calls, accept inbound calls, list and filter calls, fetch call status and analysis,
delete call data, and download per-call artifacts including mono and stereo recordings, customer/assistant
isolated recordings, video recordings, PCAP packet captures, and structured call logs.
- aid: vapi-ai:vapi-phone-numbers-api
name: Vapi Phone Numbers API
tags:
- Phone Numbers
- Telephony
- Voice AI
humanURL: https://docs.vapi.ai/api-reference/phone-numbers/list
properties:
- url: https://docs.vapi.ai/api-reference/phone-numbers/list
type: Documentation
- url: openapi/vapi-phone-numbers-api-openapi.yml
type: OpenAPI
- url: json-schema/vapi-phone-number-schema.json
type: JSONSchema
- url: json-structure/vapi-phone-number-structure.json
type: JSONStructure
- url: examples/vapi-phone-number-example.json
type: Example
- type: NaftikoCapability
url: capabilities/phone-numbers-phone-numbers.yaml
description: Buy, BYO-import, and manage phone numbers from Vapi (vapi.phoneNumber), Twilio, Telnyx, Vonage,
and SIP trunks. Attach an assistant or squad to a number for inbound routing and configure outbound caller-id
behavior.
- aid: vapi-ai:vapi-squads-api
name: Vapi Squads API
tags:
- Squads
- Voice AI
- Agents
humanURL: https://docs.vapi.ai/api-reference/squads/list
properties:
- url: https://docs.vapi.ai/api-reference/squads/list
type: Documentation
- url: openapi/vapi-squads-api-openapi.yml
type: OpenAPI
- type: NaftikoCapability
url: capabilities/squads-squads.yaml
description: Compose multiple assistants into a Squad — a multi-agent workflow where the active assistant can
hand off a live call to another assistant in the squad based on tools, intents, or transfer destinations.
- aid: vapi-ai:vapi-tools-api
name: Vapi Tools API
tags:
- Tools
- Voice AI
- Function Calling
humanURL: https://docs.vapi.ai/api-reference/tools/list
properties:
- url: https://docs.vapi.ai/api-reference/tools/list
type: Documentation
- url: openapi/vapi-tools-api-openapi.yml
type: OpenAPI
- type: NaftikoCapability
url: capabilities/tools-tools.yaml
description: Define and manage Tools (function tools, transfer-call tools, end-call tools, DTMF tools, hangup
tools, voicemail tools, MCP tools, query tools, Make/Zapier tools, knowledge-base tools, and Google/Slack/
GoHighLevel integrations) that the assistant LLM can invoke during a call.
- aid: vapi-ai:vapi-files-api
name: Vapi Files API
tags:
- Files
- Voice AI
- Knowledge Base
humanURL: https://docs.vapi.ai/api-reference/files/list
properties:
- url: https://docs.vapi.ai/api-reference/files/list
type: Documentation
- url: openapi/vapi-files-api-openapi.yml
type: OpenAPI
- type: NaftikoCapability
url: capabilities/files-files.yaml
description: Upload, list, retrieve, update, and delete files used as inputs to knowledge bases and other
assistant-side artifacts.
- aid: vapi-ai:vapi-chats-api
name: Vapi Chats API
tags:
- Chats
- Voice AI
- Text
humanURL: https://docs.vapi.ai/api-reference/chats/list
properties:
- url: https://docs.vapi.ai/api-reference/chats/list
type: Documentation
- url: openapi/vapi-chats-api-openapi.yml
type: OpenAPI
- type: NaftikoCapability
url: capabilities/chats-chats.yaml
description: Text-based conversation surface that reuses the same assistant configuration as voice calls,
including an OpenAI-compatible /chat/responses endpoint for drop-in replacement of OpenAI chat completions.
- aid: vapi-ai:vapi-sessions-api
name: Vapi Sessions API
tags:
- Sessions
- Voice AI
- State
humanURL: https://docs.vapi.ai/api-reference/sessions/list
properties:
- url: https://docs.vapi.ai/api-reference/sessions/list
type: Documentation
- url: openapi/vapi-sessions-api-openapi.yml
type: OpenAPI
- type: NaftikoCapability
url: capabilities/sessions-sessions.yaml
description: Stateful sessions that persist conversation history across calls and chats so an assistant can
pick up where the customer left off.
- aid: vapi-ai:vapi-campaigns-api
name: Vapi Campaigns API
tags:
- Campaigns
- Voice AI
- Outbound
humanURL: https://docs.vapi.ai/api-reference/campaigns/list
properties:
- url: https://docs.vapi.ai/api-reference/campaigns/list
type: Documentation
- url: openapi/vapi-campaigns-api-openapi.yml
type: OpenAPI
- type: NaftikoCapability
url: capabilities/campaigns-campaigns.yaml
description: Outbound calling campaigns — upload a customer list and schedule batched outbound calls against
one or more assistants and phone numbers with concurrency and retry policies.
- aid: vapi-ai:vapi-analytics-api
name: Vapi Analytics API
tags:
- Analytics
- Voice AI
- Reporting
humanURL: https://docs.vapi.ai/api-reference/analytics/create
properties:
- url: https://docs.vapi.ai/api-reference/analytics/create
type: Documentation
- url: openapi/vapi-analytics-api-openapi.yml
type: OpenAPI
- type: NaftikoCapability
url: capabilities/analytics-analytics.yaml
description: Run aggregate analytics queries across calls — group by assistant, time bucket, status,
end-reason, success evaluation, cost, and duration — for dashboards and chargeback reporting.
- aid: vapi-ai:vapi-insight-api
name: Vapi Insight API
tags:
- Insight
- Voice AI
- Observability
humanURL: https://docs.vapi.ai/api-reference/insights/list
properties:
- url: https://docs.vapi.ai/api-reference/insights/list
type: Documentation
- url: openapi/vapi-insight-api-openapi.yml
type: OpenAPI
- type: NaftikoCapability
url: capabilities/insight-insight.yaml
description: Define and run Insight queries across call data, including preview-before-save and on-demand
runs, to extract structured signals from conversation transcripts.
- aid: vapi-ai:vapi-observability-api
name: Vapi Observability Scorecard API
tags:
- Observability
- Scorecard
- Voice AI
- Evaluation
humanURL: https://docs.vapi.ai/api-reference/scorecards/list
properties:
- url: https://docs.vapi.ai/api-reference/scorecards/list
type: Documentation
- url: openapi/vapi-observability-api-openapi.yml
type: OpenAPI
- type: NaftikoCapability
url: capabilities/observability-observability.yaml
description: Define LLM-graded scorecards that automatically evaluate every call against your defined
success criteria and produce structured scores for trend analysis and regression detection.
- aid: vapi-ai:vapi-eval-api
name: Vapi Eval API
tags:
- Eval
- Voice AI
- Testing
humanURL: https://docs.vapi.ai/api-reference/evals/list
properties:
- url: https://docs.vapi.ai/api-reference/evals/list
type: Documentation
- url: openapi/vapi-eval-api-openapi.yml
type: OpenAPI
- type: NaftikoCapability
url: capabilities/eval-eval.yaml
description: Create test suites that simulate end-user behavior against an assistant, run them on demand,
and inspect per-run pass/fail and judge output for regression testing of voice agents.
- aid: vapi-ai:vapi-structured-outputs-api
name: Vapi Structured Outputs API
tags:
- Structured Outputs
- Voice AI
humanURL: https://docs.vapi.ai/api-reference/structured-outputs/list
properties:
- url: https://docs.vapi.ai/api-reference/structured-outputs/list
type: Documentation
- url: openapi/vapi-structured-outputs-api-openapi.yml
type: OpenAPI
- type: NaftikoCapability
url: capabilities/structured-outputs-structured-outputs.yaml
description: Define JSON-Schema-driven structured outputs that the assistant must populate from a call,
plus an on-demand /structured-output/run endpoint to backfill outputs against an existing transcript.
- aid: vapi-ai:vapi-provider-resources-api
name: Vapi Provider Resources API
tags:
- Provider Resources
- Voice AI
- Integrations
humanURL: https://docs.vapi.ai/api-reference/provider-resources/list
properties:
- url: https://docs.vapi.ai/api-reference/provider-resources/list
type: Documentation
- url: openapi/vapi-provider-resources-api-openapi.yml
type: OpenAPI
- type: NaftikoCapability
url: capabilities/provider-resources-provider-resources.yaml
description: Generic CRUD over provider-side resources (e.g. ElevenLabs voices, Cartesia voices, Deepgram
custom models) so they can be created and managed through Vapi without leaving the platform.
name: Vapi
tags:
- AI
- Voice AI
- Voice Agents
- Conversational AI
- Telephony
- Real-Time
- Transcription
- Text-to-Speech
- LLM
- Agents
- MCP
kind: contract
image: https://kinlane-productions2.s3.amazonaws.com/apis-json/apis-json-logo.jpg
access: 3rd-Party
common:
- url: https://vapi.ai
type: Portal
- url: https://docs.vapi.ai
name: Vapi Docs
type: Documentation
- url: https://api.vapi.ai/api
name: Swagger UI
type: Documentation
- url: https://api.vapi.ai/api-json
name: Raw OpenAPI JSON
type: OpenAPI
- url: https://docs.vapi.ai/quickstart
name: Quickstart
type: GettingStarted
- url: https://dashboard.vapi.ai
name: Vapi Dashboard
type: SignUp
- url: https://status.vapi.ai
name: Vapi Status
type: StatusPage
- url: https://vapi.ai/pricing
name: Vapi Pricing
type: Pricing
- url: https://discord.gg/pUFNcf2WmH
name: Vapi Discord
type: Forum
- url: https://vapi.ai/blog
name: Vapi Blog
type: Blog
- url: https://vapi.ai/changelog
name: Vapi Changelog
type: ChangeLog
- url: https://github.com/VapiAI
name: Vapi GitHub
type: GitHubOrganization
- url: https://github.com/VapiAI/server-sdk-typescript
name: TypeScript Server SDK
type: SDK
- url: https://github.com/VapiAI/server-sdk-python
name: Python Server SDK
type: SDK
- url: https://github.com/VapiAI/server-sdk-csharp
name: C# Server SDK
type: SDK
- url: https://github.com/VapiAI/server-sdk-ruby
name: Ruby Server SDK
type: SDK
- url: https://github.com/VapiAI/server-sdk-php
name: PHP Server SDK
type: SDK
- url: https://github.com/VapiAI/client-sdk-web
name: Web Client SDK
type: SDK
- url: https://github.com/VapiAI/client-sdk-python
name: Python Client SDK
type: SDK
- url: https://github.com/VapiAI/client-sdk-react-native
name: React Native Client SDK
type: SDK
- url: https://github.com/VapiAI/client-sdk-ios
name: iOS Client SDK
type: SDK
- url: https://github.com/VapiAI/client-sdk-html-script-tag
name: HTML Script Tag Client SDK
type: SDK
- url: https://github.com/VapiAI/mcp-server
name: Vapi MCP Server
type: Tool
- url: https://github.com/VapiAI/skills
name: Vapi Agent Skills
type: Tool
- url: https://github.com/VapiAI/gitops
name: Vapi GitOps
type: Tool
- url: https://github.com/VapiAI/docs
name: Vapi Docs Repo
type: CodeExamples
- url: https://docs.vapi.ai/server-url
name: Server URL (Webhooks)
type: Webhooks
- url: https://docs.vapi.ai/server-url/events
name: Server Events
type: Webhooks
- url: https://trust.vapi.ai
name: Vapi Trust Center
type: TrustCenter
- url: https://vapi.ai/privacy
type: PrivacyPolicy
- url: https://vapi.ai/terms-of-service
type: TermsOfService
- url: https://vapi.ai/aup
type: TermsOfService
- url: https://x.com/Vapi_AI
type: Twitter
- url: https://www.linkedin.com/company/vapi-ai
type: LinkedIn
- url: https://www.ycombinator.com/companies/vapi
name: Y Combinator Profile
type: Documentation
- url: plans/vapi-ai-plans-pricing.yml
type: Plans
- url: rate-limits/vapi-ai-rate-limits.yml
type: RateLimits
- url: finops/vapi-ai-finops.yml
type: FinOps
- type: Features
data:
- Sub-700ms voice-to-voice latency orchestration layer
- Transcriber + LLM + voice (TTS) modular pipeline with multi-provider choice
- Bring your own provider keys to pay providers at cost
- Assistants, Squads (multi-agent handoff), Sessions (persistent state)
- Inbound and outbound calling over Twilio, Telnyx, Vonage, vapi.phoneNumber, and SIP
- Campaigns for batched outbound calling with concurrency/retry
- Tools (function, transfer-call, end-call, DTMF, hangup, voicemail, MCP, query, Make, Zapier, GHL, KB)
- Server URL webhooks for real-time call events and function execution
- Per-call mono/stereo/customer/assistant/video recordings, PCAP packet captures, call logs
- Analytics, Insights, Observability Scorecards, and Evals for production monitoring and regression testing
- Structured Outputs for JSON-schema-constrained extraction from calls
- Chats API (text channel) and OpenAI-compatible /chat/responses
- Official SDKs - server (TS, Python, C#, Ruby, PHP) and client (Web, Python, React Native, iOS, HTML)
- Vapi MCP Server for Model Context Protocol integration
- HIPAA and Zero Data Retention add-ons; SOC2 and PCI on Scale plan
- $0.05/min Vapi orchestration fee on Build plan; model and voice costs passed at cost
sources:
- https://vapi.ai/pricing
- https://docs.vapi.ai
- https://api.vapi.ai/api
- https://github.com/VapiAI
- https://status.vapi.ai
updated: '2026-05-24'
created: '2026-05-24T00:00:00.000Z'
modified: '2026-05-24'
position: Consuming
description: Vapi is a San Francisco-based voice AI platform that lets developers build real-time, low-latency
voice agents over phone, web, and SIP. It orchestrates three modular components — a transcriber (STT), an
LLM, and a voice (TTS) — into a sub-700ms voice-to-voice pipeline, with first-class support for tools/function
calling, multi-agent squads, outbound campaigns, persistent sessions, structured outputs, recording artifacts,
evaluation scorecards, and a full REST API plus official SDKs in TypeScript, Python, C#, Ruby, PHP, and
client SDKs for Web, React Native, iOS, and Python. Vapi is a Y Combinator company.
maintainers:
- FN: Kin Lane
email: [email protected]
X: apievangelist
url: https://apievangelist.com
specificationVersion: '0.16'