Bold Agent Kit vs Vapi comes down to what you are building. Vapi is a platform developers use to build voice agents; Bold Agent Kit is the agent already built. It answers and places calls on your own Twilio number, replies to texts, email, website chat and eleven DM platforms from one knowledge base, books real calendar slots, and costs $25 a month with no per-minute charge from us.
7-day free trial · Cancel anytime
Vapi's documentation describes it as the developer platform for building voice AI agents, and its platform page says every capability is reachable through a REST API, with SDKs and a CLI alongside a dashboard. That is the right shape for a team shipping voice inside its own product. It also asks for somebody who can wire tools to your systems, choose and configure speech, model and voice providers, and look after what they built once it is live.
Bold Agent Kit is set up the way you would brief a new member of staff. Name the agent, choose a model and paste your key. Give it your documents, your Q&A pairs, your prices and hours as data points, and the pages of your site it may link to. Write its instructions: how to greet people, what to offer, and when to fetch you. Every reply on every channel, the phone included, is generated from that material and stored beside the content it used, and there is nothing to deploy.
Both products put an AI on a phone line. Vapi's docs describe calls on numbers imported from Twilio, or on free Vapi numbers, which its docs say are for US national use, up to five per account, and do not support outbound calls. Here, calls run on a number in your own Twilio account, with a separate switch for answering and for dialling, so an agent can pick up every inbound call and place none until you decide it should.
The agent hears and speaks in real time, and on a call it uses the fastest model your provider offers so it does not leave silences. Transcription comes from Google, or from Deepgram on your own key. The voice is Google, Amazon Polly or ElevenLabs, or one from your own ElevenLabs or Fish Audio library, including a voice you recorded in the browser and cloned. Every call lands in the inbox with its recording and transcript, on the caller's contact record.
Vapi's published channels are phone calls, voice calls embedded in web applications, text chat through its Chat API, and SMS chat on a 10DLC-approved Twilio number. Bold Agent Kit starts from the other end: one agent, one knowledge base, fifteen channels. Email from any mailbox, with warm-up and a health score; texts on nine carriers; calls on Twilio; a chat widget from one line of script; and Messenger, Instagram, WhatsApp and eight more DM platforms on your own app credentials.
The cookie, the email address, the phone number and the social account belonging to one person are merged into one contact, so the customer who rang on Monday and emailed on Wednesday is one row with one history, and the agent reads it before it replies. Auto-reply is a separate switch per channel, so you can start on the phone alone and switch the inbox, the website and the DMs on later.
Vapi bills its hosting by usage. At the time of writing, its published hosting rate for calls starts at $0.05 a minute, with chat and SMS charged per message, ten concurrent calls included on its Build plan and each extra line charged by the month. Model provider costs are passed on at cost, or nothing if you bring your own API key. Its Scale plan is an annual contract with a fixed platform fee, committed volume and volume-based per-minute pricing.
Bold Agent Kit is $25 a month or $200 a year for the whole account, with a 7-day free trial, unlimited agents and no per-minute, per-message or per-seat charge from us. You bring your own AI key, as you can on Vapi, and your provider bills you at cost; Twilio bills you for the number and the minutes at its rates. A month with three times the calls costs the same $25 here. The dashboard counts people rather than minutes: reached, replied, booked, and the value won.
Vapi is a serious piece of infrastructure. Its own site pitches it at builders, from individual developers to large enterprises deploying at scale, and its docs describe the job plainly: every assistant combines speech-to-text, a language model and text-to-speech, with dozens of providers to choose from, and the platform handles the infrastructure so a developer can focus on the voice experience. Assistants run on a single prompt with tools; squads orchestrate several assistants with handoffs; tools connect to your APIs and databases. If you are building a voice product, that is a strong place to start.
Most businesses comparing Bold Agent Kit with Vapi are not building a voice product. They run a business, or an agency that serves businesses, and what they want is a phone that gets answered, a calendar that gets filled and a follow-up that gets sent. For them, a developer platform means a project: somebody to build the assistant, connect it to the calendar and the CRM, decide what happens when the caller would rather text or email, and keep it all running afterwards. And the hosting bill moves with the minutes, which is usually what sends them looking for an alternative.
Bold Agent Kit is the finished agent. You build it in a dashboard from what you already have, your hours, prices, services and the answers to the questions you always get, and point it at the channels your customers use: calls and texts on your own number, email from your own mailbox, a chat widget on your site and DMs on eleven platforms. It books real open slots into one of eight calendars, reads HubSpot, GoHighLevel, Jobber, Housecall Pro, ServiceTitan or your store before it answers, and runs follow-up campaigns of email, text and call steps that stop the moment somebody replies, books or opts out.
It shares one idea with Vapi: the AI runs on your own key. You hold the account with the model provider and it bills you at cost. The difference is what sits on top. There is no hosting rate per minute and no charge per concurrent line; the plan is $25 a month or $200 a year for the account, with unlimited agents, and the only other bills are your carrier's and your AI provider's. What you give up is just as plain: it cannot transfer a live call, and it is a finished product rather than a platform to build your own software on, so a team shipping voice inside its own application should stay with a platform built for that.
None of these is a fault in Vapi. They follow from what it is: a platform for building voice agents, priced by usage. Four of them send businesses looking for something else.
Vapi's platform page makes every capability available through its REST API, and its docs point developers to SDKs and a CLI as well as a dashboard quickstart. That power is the point for an engineering team. For a clinic, a roofer or a five-person agency, it means the assistant, the tools that reach the calendar and the CRM, and the handling of every edge case are a project, and the project needs an owner long after launch day.
Hosting is billed per minute of call time and per chat or SMS message, and on the Build plan every concurrent line beyond the ten included is charged by the month. Model costs are added unless you bring your own key. That is a fair model for a product with its own revenue per call. For a business whose busiest month is also its most important, it means the phone costs most exactly when it is working hardest.
Vapi's docs describe phone calls, web calls, text chat through its API and SMS chat on a Twilio number. The enquiry that arrives by email to info@, as an Instagram DM at 10pm or through the chat bubble on your website is handled by another system with another knowledge base, or by nobody, and the same customer turns up in each of those places as a stranger.
A lead who does not pick up still needs the email on day 2 and the text on day 4. Vapi's SMS chat docs say the customer sends the first message. A sequence that emails on day 0, texts on day 2 and calls on day 5, and stops on every channel the moment the person replies, is a different kind of tool, and building one around a voice platform is a second project on top of the first.
You build the agent once, in a dashboard, from the material you already have. From then on it answers and follows up on every channel you connect, and the price does not move with the minutes.
One identity, one set of instructions, one knowledge base: documents, Q&A pairs, prices and hours as data points, and the pages of your site it may link to. Choose a model from sixteen providers on your own key, and a voice from Google, Amazon Polly or ElevenLabs, or one you cloned on your own ElevenLabs or Fish Audio key. Every reply is stored with the material it came from, so you can read back why it said what it said.
Calls on your own Twilio number, with separate switches for answering and dialling. Texts on nine carriers. Email from your own mailboxes, warmed up and health-scored. A chat widget from one line of script, and Messenger, Instagram, WhatsApp and eight more DM platforms. One person across all of them is one contact, with one history the agent reads before it replies.
It offers real open slots from your calendar and books the one the customer picks. It reads the CRM or store before it answers and writes the outcome back. Campaigns send email, text and call steps on a schedule, and stop on a reply, a booking or an opt-out on any channel. The dashboard reports people reached, replies, bookings and the value won.
This is the biggest difference between Bold Agent Kit and Vapi. Vapi is built around the voice agent, and its docs add text chat through an API and SMS chat on a Twilio number. Bold Agent Kit is built around the customer: one agent with one knowledge base that answers wherever they turn up, on the phone, by text, by email, on your website, or in Messenger, Instagram, WhatsApp, Telegram, LINE, Viber, Slack, Discord, Bluesky, Reddit or X, each connected with your own app credentials.
Because it is one agent, the answer to "are you open Saturday?" is the same on a call as in a DM, generated from the same data point. Because every handle is merged into one contact, the person who called on Monday and messaged on Instagram on Thursday is one row in the inbox, with the recording, the transcript and the DM together. Reply in any thread yourself and the agent stands down on that conversation; hand it back when you are done and it carries on where you left it.
Each one answers a problem above, and none of them needs a developer to keep it running.
The agent is written, not coded: a knowledge base, instructions and a voice. Change a price in a data point and the next caller hears the new one, with nothing to redeploy.
Nothing per minute, per message or per concurrent line from us. $25 a month for the account, with your carrier and your AI provider billing you at cost.
Calls, texts, email, website chat and DMs answered by the same agent, and one person kept as one record across all of them.
Email, text and call steps on a schedule you set, each with its own delay and condition, ended by a reply, a booking or an opt-out on any channel.
A voice platform is usually judged on its calls. Bold Agent Kit reports on people: every call, text, email, chat and booking is recorded against a contact and rolled into a funnel per agent and across all of them.
The funnel counts people rather than minutes: how many the agent reached, how many replied on any channel, how many visited your site, and how many converted, with the total value won and the median time from first contact to the booking. Goals carry a dollar value, set by a page rule or by marking a deal won, so an agent reports what it brought in rather than how long it spent on the phone.
A channel table shows where each person first made contact and which channel they were on when they converted, so you learn whether the call closes or the text after it does. One row per person shows every call with its recording, every text, email and DM, and the booking, in order. Visit triggers close the loop: when a known caller opens your pricing page and leaves, the agent tells you, or follows up after a delay on the channel they last used.
Built from the cookie, email address and phone number attached to that contact, so activity months apart on different channels appears on one row.
Page goals and visit triggers require a one-line snippet on your website. Conversations, bookings and manually marking a deal won work without it.
What each product does for the jobs a business actually buys a voice agent for. The Vapi column is taken from Vapi's own website and documentation on the date shown below.
| The job | Vapi | Bold Agent Kit |
|---|---|---|
| Building the agent | A dashboard, plus a REST API that exposes every capability, with SDKs and a CLI for developers | A dashboard: knowledge base, instructions, voice and channels, with no code |
| Answering and placing phone calls | Calls on numbers imported from Twilio, or on free US numbers that do not support outbound calls | Inbound and outbound on your own Twilio number, with a switch for each |
| Text messages | SMS chat on a 10DLC-approved Twilio number, started by the customer | Replies and campaign texts on nine carriers, inside the hours you set |
| Email, website chat and social DMs | Voice calls embedded in web apps, and text chat through its Chat API | Your own mailboxes with warm-up, a chat widget and eleven DM platforms |
| Booking and the CRM | Tools that call your own APIs, and a library of integrations that includes Google Calendar, HubSpot and Salesforce | Eighteen native: eight calendars, HubSpot, GoHighLevel, Jobber, Housecall Pro, ServiceTitan and five stores |
| The AI model | Your choice of providers; model costs passed on at cost, or nothing on your own API key | Your own key with one of sixteen providers, billed by them at cost |
| Call history | Kept for 14 days on the Build plan, custom on Scale | Recordings and transcripts kept on the contact record |
| The platform charge | Hosting billed per minute of calls and per chat or SMS message, with extra concurrent lines charged monthly | $25 a month or $200 a year for the account, nothing per minute or message |
Vapi is a developer platform for building voice agents, with a dashboard, an API, multi-assistant squads and a wide choice of speech, model and voice providers, billed by usage or on an annual contract. Bold Agent Kit is a finished agent for businesses and agencies that works calls, texts, email, website chat and DMs from one knowledge base, at $25 a month or $200 a year for the account. Both let you bring your own AI key. If you are building voice into your own product, Vapi is made for that; if you want the phone answered and the follow-up done without writing code, Bold Agent Kit is.
We checked Vapi’s public website on 11 September 2026 for this comparison. Plans and features change, so check vapi.ai for the current detail. Vapi is a trademark of its owner, and Bold Agent Kit is not affiliated with or endorsed by it.
Most people arrive at this comparison in one of three positions. Here is what each one looks like in practice.
You built, or paid somebody to build, a Vapi assistant that answers your business line. Now you want the same calls answered without the upkeep, with texts, email and DMs handled by the same agent, and without a hosting bill that rises with the minutes. You rebuild the knowledge base from the prompt and material you already have, connect your Twilio number, and switch answering on.
An agency or product team that builds custom voice applications for larger clients on Vapi, and needs something quicker for the local businesses that simply want their phone, inbox and DMs answered. Keep the developer work on Vapi, and run the small-business accounts here: one agent per client, each with its own number, knowledge base, voice and client login, on one $25 plan.
Choose Vapi if you are a developer or an engineering team building voice into your own product or platform, and you want control of every layer: the speech-to-text, model and voice providers, squads of assistants with handoffs, custom SIP, warm transfers, tools that call your own APIs, and a REST API for everything. Its Scale plan adds an annual contract with options such as SSO, RBAC and data residency. Bold Agent Kit is not built for that job, and pretending otherwise would waste your time.
Connect the tools you already run and the agent works them like an assistant would: it takes appointments straight into your calendar, writes notes and outcomes back to the CRM, and reads your store to see what somebody has bought before it answers them.
Bookings are real, not guesses. It is only ever given slots that are genuinely free, and it can only take one it actually offered — so it cannot invent a time or double-book you. The same connection pulls a segment of past customers into a list when you want to run a campaign at them.
HubSpot
GoHighLevel
Housecall Pro
ServiceTitan
Jobber
Google Calendar
Microsoft 365
Calendly
HubSpot Meetings
Zoho Bookings
Acuity
Cal.com
Square
Stripe
Shopify
WooCommerce
Wix
SquarespaceEach agent uses a model on your own API key, in the same way it sends from your own mailbox and your own phone number. Sixteen providers are supported: OpenAI, Anthropic, Gemini, xAI and Mistral; DeepSeek, GLM, Kimi, Qwen and MiniMax; Groq, Together, Fireworks and OpenRouter; and Ollama or any other OpenAI-compatible endpoint you host yourself.
Bring your own credentials and you own the whole stack: your model provider, your mailboxes, your carrier, your CRM. We charge for the software and nothing else — no token markup, no metering, no credit packs — so your provider bills you direct at their rates, usually a few dollars a month per agent.
OpenAI
Anthropic
Gemini
xAI
Mistral
DeepSeek
GLM
Kimi
Qwen
MiniMax
Groq
Together
Fireworks
OpenRouter
Ollama
Any OpenAI APIA new address sending forty cold emails a day gets filtered, and takes the domain with it. Mailboxes that opt into warm-up send each other ordinary mail on a ramp, starting at five a day. The receiving mailbox finds each message, moves it out of spam if it landed there, marks it read and replies to a share of them.
Each mailbox carries a health score out of 100 — SPF, DKIM and DMARC on the sending domain, seven days of real sends and bounces, and where its own warm-up mail is landing — with the reasons listed. A mailbox whose bounce rate or spam placement passes a threshold is pulled out of campaign rotation automatically, before it damages the domain.
Three steps. There is nothing to deploy, and nothing dials or sends until you switch it on.
Open your Vapi assistant's system prompt next to a new agent here. Name the agent, choose a model and paste your API key; on calls it switches to your provider's fastest model on its own. Then take the prompt apart: the facts in it, prices, hours and lead times, become data points; the answers it was told to give become Q&A pairs; anything longer becomes a document. The rules it followed become its instructions.
Choose the voice: Google, Amazon Polly or ElevenLabs, or connect your own ElevenLabs or Fish Audio account and use a voice from your library, including one you cloned. If your assistant also made outbound calls, save its script here as a call template, ready to use as a step in a campaign.
Add your Twilio account and the number your customers already call. A number held in your own Twilio account runs here; if the number was rented through a platform rather than held in Twilio, check with that platform what moving it involves, or start on a new Twilio number. Switch answering on, and leave dialling off until you want outbound calls.
Connect the calendar it books into and the CRM or store that holds your customers, then the channels you did not have before: a mailbox, the chat snippet, your social accounts. Import your contacts as a CSV, or pull a segment straight out of the connected CRM.
Call the number yourself before you move any traffic. Ask for your hours, a price and a Thursday slot, and read the transcript in the inbox a minute later. Adjust a data point or an instruction and ring again. When it answers the way your Vapi assistant did, point your customers at it.
Then add what the phone alone was not doing: a campaign of email, text and call steps for the leads who did not book, created as a draft, reviewed on one screen, and started only when you press Start.
One account runs unlimited agents, one per client, and nothing is shared between them. A knowledge base belongs to a single agent, and content from one agent is never used to generate replies for another.
Invite a client by email address and they get their own login, which opens onto the one agent you gave them: its conversations, its numbers, its analytics and a reply box. They cannot see your other clients, your API keys, your settings or your billing. Access is deny-by-default.
No seat tiers, no credit packs, no per-agent or per-channel pricing, and nothing held back behind an enterprise call.
| Agents | |
|---|---|
| How many | Unlimited — one per business, brand or job |
| Shared brain | The same identity, instructions and knowledge on every channel |
| Knowledge base | Documents, Q&A pairs, data points and pages of your own site |
| Isolation | One agent's content is never used to reply for another |
| Client seats | Invite a client into one agent, with their own login |
| Channels | |
| As many mailboxes per agent as you connect, rotated by campaigns | |
| Text messages | Nine providers — Twilio, Telnyx, SimpleTexting, Plivo, Vonage, Sinch, Bandwidth, MessageBird, ClickSend |
| Phone calls | Answering and dialling, with recordings and transcripts |
| Transcription | Google, or Deepgram nova-3 on your own key |
| Voices | Google, Amazon Polly and ElevenLabs, or your own ElevenLabs / Fish Audio account |
| Voice cloning | Record a voice in the browser and clone it on your ElevenLabs or Fish Audio key |
| Website chat | A widget on your site with your colour, image and greeting |
| Social DMs | Messenger, Instagram, WhatsApp, Telegram, LINE, Viber, Slack, Discord, Bluesky, Reddit, X |
| Inbox | |
| One list | Every channel's conversations together, newest first, tagged by channel |
| Filters | By channel, by mailbox or number, or by agent |
| Takeover | Reply yourself and the agent stands down on that thread |
| Hand back | Return the thread to the agent when you are done |
| Campaigns | |
| Multichannel sequences | Email, text and call steps in one campaign, each with its own timing |
| Send conditions | Send a step only if they opened, did not open, clicked or did not click |
| Schedule | The days, hours and time zone anything may be sent |
| Stop rules | A reply, a booking or an opt-out on any channel ends the sequence |
| Pacing | Per-mailbox and per-number hourly limits, pausable and resumable |
| Review before start | The full plan is shown; a campaign sends nothing until you start it |
| Templates | Email, text and call scripts, saved once and reused |
| Analytics | |
| Funnel | Reached, replied, visited, converted, plus value won and time to convert |
| By channel | Where each person started, and the channel they converted on |
| Pages | Which pages known people read, and how many of them converted |
| Journeys | One row per person: every message, visit, call and booking in order |
| Goals | A page rule or a manual "won", each with a dollar value |
| Triggers | A known person returns to a page — notify you, or send a follow-up |
| Open and click tracking | Recorded per campaign and added to that person's history |
| Integrations | |
| Calendars | Google, Microsoft 365, Calendly, Cal.com, Acuity, Zoho, Square, HubSpot Meetings |
| CRMs and commerce | HubSpot, GoHighLevel, Housecall Pro, ServiceTitan, Jobber, Stripe, Shopify, WooCommerce, Wix, Squarespace |
| Booking | Real open slots offered, and only an offered slot can be booked |
| CRM import | Import a segment of customers into a contact list |
| Deliverability and compliance | |
| Mailbox warm-up | Ramped sending across the network, with spam rescue and replies |
| Health score | SPF, DKIM, DMARC, bounce rate and inbox placement, with reasons |
| Auto-rest | A failing address is removed from rotation automatically |
| Unsubscribes | One click, honoured automatically, including on queued mail |
| Reply limits | One reply per thread, per-sender daily and per-channel hourly caps |
| Filtering | Bounces, out-of-office and no-reply addresses ignored |
| List building | |
| Contact extraction | Emails, phone numbers and social profiles, on any site. Unlimited. |
| Deep scan | Follows the pages of a domain most likely to carry contact details |
| Domain queue | Load a CSV of sites and work through them one at a time |
| Contacts and lists | Unlimited, by email or phone, with CSV import and export |
| AI | |
| Your own key | Sixteen providers, including any OpenAI-compatible endpoint you host |
| What we charge for AI | Nothing — no markup, no metering, no credit packs |
| On calls | Uses your provider's fastest model automatically |
| Account | |
| Platforms | Chrome side panel and web dashboard on one login |
| Members' forum | Six boards, visible only once you are signed in |
| Referral program | $50 a year per referral ($5 a month), recurring for the life of the account |
| Support | support@boldagents.co |
Billed monthly. No add-ons, no seat tiers. Cancel anytime.
$16.67 a month, billed once a year. No add-ons, no seat tiers. Cancel anytime.
It depends on who is buying. For a business, or an agency serving businesses, that wants its phone, texts, email, website chat and DMs answered and followed up without writing code, Bold Agent Kit is the better fit: it is a finished agent, it books real calendar slots, and it costs $25 a month for the account with no per-minute charge from us. For a developer or an engineering team building voice into its own product, wanting squads of assistants, custom SIP, warm transfers and control of every provider, Vapi is the better fit. Neither is better in general; they are built for different buyers.
Vapi is a developer platform for building voice agents: you assemble speech-to-text, a language model and a voice from a wide choice of providers, connect tools to your own systems, and reach everything through its dashboard or its REST API. It bills hosting by usage. Bold Agent Kit is the agent already built for a business: one knowledge base answering calls, texts, email, website chat and social DMs, with booking into eight calendars, native CRM and store connections, follow-up campaigns and one inbox that holds every channel. It is one flat plan. Both let you run the AI on your own key.
Vapi prices by usage. At the time of writing, its published hosting rate for calls starts at $0.05 a minute, chat and SMS are charged per message, and its Build plan includes ten concurrent calls with extra lines charged monthly. Model costs are passed on at cost, or nothing on your own key, and its Scale plan is an annual contract with a platform fee and committed volume. Bold Agent Kit is $25 a month or $200 a year for the account, with unlimited agents and no per-minute, per-message or per-line charge. Twilio bills you for the number and minutes, and your AI provider for the model, both at cost.
You rebuild rather than import, and it usually fits into an afternoon. Take your assistant's system prompt apart into a knowledge base: facts become data points, scripted answers become Q&A pairs, longer material becomes documents, and the rules become instructions. Connect your Twilio account and number, your calendar and your CRM, then import your contacts as a CSV or as a segment from the connected CRM. Ring the number yourself, read the transcript, adjust, and switch answering on when it sounds right. A number rented through another platform, rather than held in your own Twilio account, may need moving first.
When you are building voice into your own product. Vapi is made for developers and engineering teams who want control over each layer: which speech-to-text, model and voice providers to use, squads of assistants that hand calls to each other, tools that call your own APIs, custom SIP and warm transfers, and a REST API for every capability. Organisations that want an annual contract with SSO, RBAC and data residency will find those on its Scale plan. If that describes you, Vapi fits better than a finished business agent, and Bold Agent Kit would get in your way.
Yes. Every agent runs on a model on your own API key, and sixteen providers are supported: OpenAI, Anthropic, Gemini, xAI and Mistral; DeepSeek, GLM, Kimi, Qwen and MiniMax; Groq, Together, Fireworks and OpenRouter; and Ollama or any other OpenAI-compatible endpoint you host yourself. Your provider bills you directly at its rates, with nothing metered, marked up or sold as credits on our side. On a call, the agent switches to the fastest model your provider offers automatically. Voices can run on your own ElevenLabs or Fish Audio key, and transcription on your own Deepgram key.
No. Everything is set up in the dashboard. You name the agent, paste your AI key, and give it your material: documents, Q&A pairs, prices and hours as data points, and the pages of your site it may link to. Channels connect with credentials you already have: a Twilio account for calls, an app password or SMTP and IMAP details for email, one line of script for the website chat, and your own app credentials for each social platform. Calendars and CRMs are chosen from a list of eighteen. An agent can be answering your phone the same afternoon.
No. It cannot put a live call through to another number, whereas Vapi's platform page lists warm transfers. What it does instead is what a good receptionist does when you are busy: it takes the caller's name, number and reason for calling, tells them what will happen next, and puts the recording and transcript in your inbox, tagged as a call, on their contact record. You name in its instructions which calls get that treatment. If a live handover to your team is essential to how you work, that is a real reason to choose a platform that does it.
How Bold Agent Kit compares with the other tools people shortlist for the same job.
Rebuild the assistant as a knowledge base, connect your Twilio number and ring it yourself. Then put email, texts and DMs on the same agent, for $25 a month.
7-day free trial · Cancel anytime