How much does Vybe cost?

Free Starter with one builder and 100 daily credits. Pro is $49 per builder per month with 2,000 credits per builder, extra credits at $0.01 each and end user seats at $7 a month. Team is $99 per builder with a 3 seat minimum, 15,000 shared credits plus 5,000 per extra builder, end user seats at $12, granular permissions and SSH tunneling. Enterprise is custom with unlimited credits.

Does Vybe have SOC 2, SSO and role based access control?

Full under the Agentic Index verification standard. Vybe publishes public evidence of enterprise security and identity controls, covering a compliance posture such as SOC 2 or ISO 27001, single sign on (SSO or SAML), and role based access control (RBAC). Evidence basis: Vybe documents both compliance and access controls. It is SOC 2 Type II certified with regular third party penetration testing, encryption in transit and at rest, and workspace isolation. Every app sits behind authentication, members have r. Graded from public evidence only against the Agentic Index 14 point capability taxonomy. No vendor pays for placement.

Can you bring your own model or route between providers with Vybe?

Full under the Agentic Index verification standard. Vybe publishes public evidence of bring your own model and bring your own key (BYOK), and of routing tasks between models or providers rather than being locked to one. Evidence basis: The customer chooses the model per message. A selector in the chat input switches between Claude Opus 4.5, Claude Sonnet 4.5, Claude Haiku 4.5, GPT-5.1 and Gemini 3 Pro at any time, each with a documented relative credit cost. Graded from public evidence only against the Agentic Index 14 point capability taxonomy. No vendor pays for placement.

Can Vybe be self hosted or run in a private VPC?

Partial under the Agentic Index verification standard. Vybe publishes some public evidence on deployment and data residency, but narrower than a documented self hosted or private VPC option. Evidence basis: Vybe is vendor hosted only. Apps deploy to a Vybe production URL, and no self hosted, private cloud or region choice is documented. The Team plan adds SSH tunneling to reach private databases, which keeps data sources inside the customer's. Graded from public evidence only against the Agentic Index 14 point capability taxonomy. No vendor pays for placement.

Back to vendors

Vybe

Q: Who is Vybe for?

operations teams, finance teams, GTM teams, startups, mid-market

Also known as: Vybe, vybe.build, Vybe AI, Vybe internal tools

Visit site

Coding agentindependentVerified 2026-10-02

Entry price$49/builder/mo, Pro · free planFull pricing detail

AI platform for internal tools whose agent writes full stack Next.js apps with their own database and runs actions across 3,000+ connected services from chat, with per message model choice, recipes and SOC 2 Type II.

Vybe is an AI platform for building internal tools only. A builder describes the app, and Vybe's agent writes a full stack Next.js application with its own PostgreSQL database, API routes, integration connections and production deployment. The agent creates tables, writes queries and manages migrations as it goes, and the builder can open every file in a Code tab or restore any automatic checkpoint.

The agent also acts directly in connected services during a conversation. Asked to create a Slack channel, open a Jira issue or send the team an email through Gmail, it picks the integration and action itself and runs it, and it can chain several services in one request. It works with any integration the organization or the individual has connected, across more than 3,000 services, and prompts the user to authorize a service it needs that is not yet connected.

Builders pick the model per message, from Claude Opus, Sonnet and Haiku 4.5 to GPT-5.1 and Gemini 3 Pro. Recipes and templates give starting points with integrations already wired, and custom instructions keep coding conventions and domain rules in front of the agent. Vybe is SOC 2 Type II with regular penetration testing, role based members and per app access levels.

The agent runs integration actions without a documented approval step, works only when prompted, and lives entirely on Vybe's cloud. Vybe's marketing describes agents that operate apps after they are built, but the documentation covers the building agent and Inngest workflows, not standing operating agents.

Vendor details

Canonical URL

https://www.vybe.build

Subcategory

AI app builder for internal tools

Funding status

Private (Vybe, Inc.). Funding details are not established from the company's own pages.

Company status

independent

Use cases & customers

Primary use cases

prompt to app generation of internal toolsadmin panels and CRMs over live company datarunning integration actions from chat (Slack, Jira, Gmail)scheduled reporting toolsinternal dashboards for ops, finance and GTM teams

Target customers

operations teamsfinance teamsGTM teamsstartupsmid-market

Deployment options

Vybe cloud (vendor hosted)SSH tunneling to private databases (Team)

Integrations

The agent reaches 3,000+ services through Vybe's integration platform, including Jira, Slack, Gmail, Stripe, Salesforce, HubSpot, Google Ads and Linear, and runs their search, list, read, create, update and execute actions from chat. Apps connect to PostgreSQL, MySQL and Amazon Redshift, custom APIs with stored credentials, and typed clients generated from OpenAPI specs.

In practice

An ops lead asks Vybe for a refunds admin panel over the company's Postgres and Stripe data, then deploys it to the team.

A manager asks the agent in chat to pull open Jira tickets, find the assignees in Slack and send each one a reminder.

A RevOps analyst starts from a recipe with Salesforce already wired and asks for a weekly pipeline report as a scheduled workflow.

Sources & related URLs

Related / legacy domains

https://docs.vybe.build https://www.vybe.build/pricing.json https://www.vybe.build/security

Research sources

https://docs.vybe.build/ https://docs.vybe.build/getting-started/faq https://docs.vybe.build/integrations/using-in-chat.md https://docs.vybe.build/getting-started/plans-and-credits.md https://docs.vybe.build/security/overview.md https://www.vybe.build/pricing https://docs.vybe.build/llms.txt

Agentic Index coverage score

8.0 / 14 capabilities · 57%

Integrations & Tool Calling	Full
The agent takes authenticated action in outside systems. During a chat it identifies the relevant integration, selects an action and executes it, with documented examples of creating a Slack channel, opening a Jira issue and sending email through Gmail. Its search, list, read, create, update and execute capabilities span 3,000+ services. It also creates tables and manages migrations in the app database. Sourcedocs.vybe.build/integrations/using-in-chat.mdread 2026-10-02
Workflow Orchestration	Partial
The agent runs multi step work on its own. It chains several integrations in one request (pull Jira tickets, find assignees in Slack, send reminders) and builds apps across pages, tables and routes. No multi agent execution is documented, and Inngest workflows run generated app code rather than the agent. Sourcedocs.vybe.build/integrations/using-in-chat.mdread 2026-10-02
Knowledge Grounding & RAG	Partial
The agent grounds its work in live data rather than a retrieval layer. It accesses data through connected integrations and the app's PostgreSQL database, and saved named queries can be reused across apps. No maintained knowledge base, index or embeddings store over the customer's documents is documented. Sourcedocs.vybe.build/getting-started/faqread 2026-10-02
Human Oversight & Guardrails	Partial
Oversight comes from recovery and scoping rather than approval before an action runs. The integrations page describes the agent executing actions such as sending email directly, with no approve step first. Control comes from the agent's two operating modes, integration visibility limited to what the user can access, prompts to authorize unconnected services, and checkpoint rollback of app state. Sourcedocs.vybe.build/integrations/using-in-chat.mdread 2026-10-02
Security, Identity & Governance	Full
Vybe documents both compliance and access controls. It is SOC 2 Type II certified with regular third party penetration testing, encryption in transit and at rest, and workspace isolation. Every app sits behind authentication, members have role based permissions, apps carry organization, restricted or user level access, and the Team plan adds granular permissions. Sourcedocs.vybe.build/security/overview.mdread 2026-10-02
Observability & Auditability	Partial
Vybe keeps a partial record of what happens. The editor has a logs tab, workflows can be monitored from a Workflows tab, checkpoints record app states, and integration action results appear in the chat thread. No retained audit trail of the agent's actions in outside systems is documented, which keeps it short of Full. Sourcedocs.vybe.build/llms.txtread 2026-10-02
Memory & State Persistence	Partial
The agent keeps conversation state within a thread, with automatic checkpoints of app state. Custom instructions give the agent persistent coding conventions, design guidelines and domain rules, which are authored instructions rather than memory. No memory layer with a stated scope and lifetime is documented. Sourcedocs.vybe.build/llms.txtread 2026-10-02
Deployment & Data Residency	Partial
Vybe is vendor hosted only. Apps deploy to a Vybe production URL, and no self hosted, private cloud or region choice is documented. The Team plan adds SSH tunneling to reach private databases, which keeps data sources inside the customer's network but not the platform. Sourcevybe.build/pricingread 2026-10-02
Prebuilt Agents, Templates & Packs	Full
Buyers get adoptable starting points for the agent's work. Recipes are pre configured app starters with guided setup steps and built in integrations, and Templates are pre built app blueprints from the Vybe team and community. The Creating an App page documents starting from either. Sourcedocs.vybe.build/llms.txtread 2026-10-02
Triggers & Channel Coverage	Partial
In the documentation, work reaches the agent only through chat threads in the app builder. Scheduled jobs and event driven functions run as Inngest workflows the agent writes into the app, and apps cannot receive inbound webhooks directly. The pricing page describes agents that operate apps on schedules, but no documentation page covers such agents. Sourcedocs.vybe.build/llms.txtread 2026-10-02
Model Flexibility & Routing	Full
The customer chooses the model per message. A selector in the chat input switches between Claude Opus 4.5, Claude Sonnet 4.5, Claude Haiku 4.5, GPT-5.1 and Gemini 3 Pro at any time, each with a documented relative credit cost. Sourcedocs.vybe.build/getting-started/plans-and-credits.mdread 2026-10-02
APIs, SDKs & MCP Extensibility	Unable to verify
No documented interface to Vybe's own platform exists. No platform API, SDK or MCP server is listed in the docs index, and apps cannot receive inbound webhooks directly. Custom APIs and OpenAPI client generation let generated apps call other services, but that is outbound integration rather than access to Vybe itself. Sourcedocs.vybe.build/llms.txtread 2026-10-02
Testing, Debugging & Optimization	Partial
Vybe offers debugging support rather than evaluation. The agent explains and debugs on request, a live sandbox previews changes with hot reload, troubleshooting and platform limitation guides exist, and workflows can be tested from the Workflows tab. No stored, scored test cases for the agent are documented. Sourcedocs.vybe.build/llms.txtread 2026-10-02
Browser & Computer Use	Unable to verify
No browser, desktop or remote computer control by the agent is documented. It acts through integration APIs and generated code. Sourcedocs.vybe.build/llms.txtread 2026-10-02

The Agentic Index coverage score grades every vendor Full, Partial or Unable to verify against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded

Recent platform changes

2026-09-28·Agent capabilityVerified

Vybe agents now keep track of unfinished work in a thread and check back in on their own, without being prompted. Over time they also learn how each person works, so their follow ups read closer to what that person would write.

Bears on: Memory / state

View source

View all 1 change for Vybe →Tracked since Sep 2026 · Verified from public vendor sources

Pricing

$49/builder/mo, Pro · free plan

builder seats with included credits, end user seats, and metered extra credits

Free tierTrial available

Included quota

Pro: 2,000 credits per builder a month. Team: 15,000 shared credits a month plus 5,000 per additional builder. Free: 100 credits a day up to a 2,000 balance.

What is public

Builder and end user seat prices, included credits, extra credit price and plan features are published, including a machine readable pricing.json. Enterprise is custom.

Billing mechanics

Monthly billing per builder seat plus end user seats, with a credit allowance per builder and prorated seat changes.

Cost watchouts

Every agent interaction consumes credits, and stronger models cost more per message. End users who open deployed apps are billed per seat ($7 on Pro, $12 on Team). Plan credits do not roll over, though purchased credits do.

Variable cost rationale

Seat prices are fixed, but credits are metered per message at model dependent rates with a published top up price, and end user seats grow with adoption.

Additional watchouts

Budget builder seats, end user seats and credits separately; at rollout, end user seats often become the largest line.

Overage / add-ons

Additional credits are purchased at $0.01 each on Pro and Team and do not expire; on Free, AI work pauses until the daily reset.

Sales call required

Mixed (some tiers require a call)

Free / trial

Free Starter forever: 1 builder, 100 daily credits, 1,000 credit bonus, unlimited apps and integrations

Lowest paid plan

Pro at $49 per builder per month

Commercial notes

Pro is self serve; Team routes to sales by email and adds a dedicated CSM and Slack channel; Enterprise adds a forward deployed engineer and custom SLAs.

Key ambiguities

The docs list Team at 5,000 credits a month while the pricing page lists 15,000 shared credits for the three seat minimum. The pricing page figures are shown here, so confirm the Team allowance with the vendor.

Missing data

Enterprise pricing and annual discounts.

Agentic Index verified 2026-10-02

Official pricing page Docs

Data confidence: medium