Vapi Web SDK review
Vapi Web SDK is a TypeScript npm package that lets you start Vapi voice assistant calls directly in your web app.
TL;DR
Vapi Web SDK is a browser package for starting calls with Vapi voice assistants from a web app. It is for developers adding voice AI to web pages. It handles call control, microphone muting, playback volume, text messages to the assistant, recording and call events.
- 70GitHub stars
- +1new stars, week of Oct 25
- ~319.8kdownloads a week, +3% vs 4-week avg
- #17in Voice and realtime this week
Vapi Web SDK at a glance
- Category
- Voice and realtime
- Maintained by
- Vapi (organization, @VapiAI on GitHub)
- Language
- TypeScript, Shell, JavaScript
- Repository
- github.com/VapiAI/client-sdk-web
- npm package
- @vapi-ai/web 3.0.0
- Latest release
- v3.0.0 on GitHub, Oct 16, 2026
- First published
- Sep 14, 2023
What Vapi Web SDK does
- Vapi Web SDK starts a call from an assistant ID or an inline assistant object with model and voice settings.
- Vapi Web SDK overrides assistant parameters per call, such as setting template variables or disabling recording, through assistant overrides.
- Vapi Web SDK sends text messages to the assistant during a call with roles such as system, user, assistant, tool or function.
- Vapi Web SDK mutes and unmutes the user's microphone and sets assistant playback volume from 0 to 1.
- Vapi Web SDK can make the assistant speak a message and optionally end the call afterward.
- Vapi Web SDK emits events for speech start and end, call start and end, volume levels, audio, messages, transcripts and errors.
- Vapi Web SDK can start and stop video recording manually, including retrying after a recording failure.
How it works
You install the @vapi-ai/web package, import the Vapi class and create an instance with your Vapi Public Key. Calling start begins a call with a Vapi assistant. The SDK uses Daily (@daily-co/daily-js) for the call connection and recording. It exposes the audio element that plays the assistant, so you can connect it to tools like a waveform visualizer.
How to install and start using Vapi Web SDK
With npm
npm install @vapi-ai/webRequirements
- Node.js >= 22.14.0 to install version 3.0.0.
- A Vapi Public Key passed to the Vapi constructor.
First steps
- Import the class with import Vapi from '@vapi-ai/web'.
- Create an instance by passing your Vapi Public Key to the Vapi constructor.
- Call vapi.start with your assistant ID or an assistant object to begin a call.
- Call vapi.stop() to stop the recording and close the connection.
Who Vapi Web SDK is for
- Web developers adding voice assistant calls to a browser app.
- Apps that need custom call UI driven by speech, volume and transcript events.
Limitations to know
- Vapi Web SDK starts calls with Vapi assistants, so it depends on the Vapi service and a Vapi Public Key.
- Version 3 requires Node.js 22.14.0 or later; projects that cannot upgrade must stay on @vapi-ai/web@^2.
- Browser autoplay policies can block assistant audio when a call is not started from a user gesture.
- A failed recording start is not retried automatically; you must call startRecording() yourself.
Is Vapi Web SDK growing? Usage data
In the week of Oct 25–31, 2026, Vapi Web SDK ranked #352 on the Devtoolwatch Board and #17 in Voice and realtime, with a momentum score of 21 out of 100. Its main package had about 319.8k downloads that week, +3% against its own 4-week average.
Show the weekly numbers
| Week of | New stars |
|---|---|
| Aug 9, 2026 | 0 |
| Aug 16, 2026 | 1 |
| Aug 23, 2026 | 1 |
| Aug 30, 2026 | 0 |
| Sep 6, 2026 | 1 |
| Sep 13, 2026 | 0 |
| Sep 20, 2026 | 0 |
| Sep 27, 2026 | 1 |
| Oct 4, 2026 | 1 |
| Oct 11, 2026 | 0 |
| Oct 18, 2026 | 0 |
| Oct 25, 2026 | 1 |
The Board ranks tools by four signals each week: new stars, new stars for the repo's size, added downloads and download growth beyond the market. How the ranking works.
Vapi Web SDK alternatives
Other voice and realtime AI tools on the Board in the week of Oct 25–31, 2026, by momentum:
| Rank | Tool | Stars | Downloads/wk |
|---|---|---|---|
| #1 | faster-whisperfaster-whisper is a Python reimplementation of OpenAI's Whisper speech-to-text model built on the CTranslate2 inference engine for… | 26.4k | 4.4M |
| #2 | KokoroKokoro is a Python inference library for Kokoro-82M, an open-weight text-to-speech model with 82 million parameters. | 9.8k | 150.7k |
| #3 | ChatterboxChatterbox is a family of open-source text-to-speech and voice cloning models by Resemble AI, with English and multilingual variants. | 27.5k | 73.2k |
| #4 | VoiceMemInfrastructure for the next generation of voice agents, designed to provide universal memory. It is divided into a left brain and a… | 2.8k | – |
| #5 | PipecatPipecat is an open-source Python framework for building real-time voice and multimodal conversational agents, maintained by Daily and… | 17k | 348.5k |
| #6 | WhisperXWhisperX is a Python speech recognition tool built on Whisper that adds word-level timestamps and speaker diarization. | 25k | 158.6k |
Recent Vapi Web SDK releases
Vapi Web SDK FAQ
Is Vapi Web SDK free and open source?
Vapi Web SDK's code is on GitHub (VapiAI/client-sdk-web), but the repository doesn't declare a standard open-source license. Check its license file before using it commercially.
How do I install Vapi Web SDK?
Run npm install @vapi-ai/web. It needs Node.js >=22.14.0.
Is Vapi Web SDK still maintained?
Yes. The VapiAI/client-sdk-web repository was last updated on Oct 16, 2026, and the latest release, v3.0.0 on GitHub, came out on Oct 16, 2026.
How popular is Vapi Web SDK?
Vapi Web SDK has 70 GitHub stars as of Nov 1, 2026. It gained 1 new stars in the week of Oct 25–31, 2026, against a 12-week average of 1 a week. Its main package had about 319.8k downloads that week, +3% against its own 4-week average. That made it #17 among voice and realtime AI tools on the Devtoolwatch Board.
What are the best alternatives to Vapi Web SDK?
Other voice and realtime AI tools on the Devtoolwatch Board in the week of Oct 25–31, 2026, by momentum: faster-whisper (#1), Kokoro (#2), Chatterbox (#3), VoiceMem (#4), Pipecat (#5). The alternatives table above has their numbers.
How do I start a call with Vapi Web SDK?
Vapi Web SDK starts a call when you call vapi.start. Pass either an assistant ID or an assistant object with model and voice settings. You can add an overrides object to set variables like a name in the first message.
Why is there no assistant audio in my Vapi Web SDK call?
Vapi Web SDK cannot play audio when the browser's autoplay policy blocks playback, usually because the call was not started from a user gesture. The SDK emits an error with type audio-start-failed, which you can use to prompt the user to enable audio.
Does setVolume in Vapi Web SDK change what the assistant hears?
No. Vapi Web SDK's setVolume affects playback only, from 0 to 1. It does not change the assistant's input or volume-level event values. The setting persists across calls and can be set before a call starts.
How does recording work in Vapi Web SDK?
Vapi Web SDK records automatically when video recording is enabled in the call's artifact plan. Success fires recording-started and failure fires recording-error. Failures are not retried, but you can call startRecording() and stopRecording() yourself.
Did Vapi Web SDK 3.0.0 change the API?
No. Vapi Web SDK 3.0.0 made no SDK API changes. It upgraded @daily-co/daily-js to 0.93.0 and now requires Node.js 22.14.0 or later to install.
Sources
Vapi Web SDK README badge
Maintainers can show the tool's weekly Board rank in their README:
[](https://index.devtoolwatch.com/tools/vapi/)How this page is made: Devtoolwatch writes each review from the project's README, releases, package metadata and website, with an AI model, then checks every claim against the source it cites and drops what the sources don't support. The numbers come from Devtoolwatch's daily GitHub snapshots, GitHub's stargazer history, and the npm and PyPI download APIs. Spotted an error? Email [email protected].