Home / Run Operations & Workflows / MindStudio

MindStudio

A visual IDE for building your own AI agents and apps — 200-plus models at cost, deployed as web apps, automations, extensions or MCP servers.

Solo Founder AgentsChecked
Solo-fit verdictSolo-Ready

The builder's pick for founders who want to own their agents and their model bill — the catch is usage is metered separately on every plan.

True solo cost · low$16/mo$16/month billed annually (Individual) plus at-cost usage — the platform fee for unlimited agents and runs; light agents on small models add only a few dollars of usage in a typical month.
True solo cost · typical$16/mo$16/month billed annually (Individual) plus usage at provider cost — usage, not the platform fee, is the variable: heavy agents on frontier models can add tens of dollars, and workspace budgets can cap it.
Rank #14 of 16 in OperationsFree start Free plan availableSetup An afternoon for a first working agent; days to feel fluent in the editorControl Acts + logs everythingNo-code 4/5Checked · prices checked 2026-11-06

Visit MindStudio ↗ Compare all run operations & workflows agents

The short version

As of November 2026, MindStudio costs $16/month billed annually for unlimited agents, plus model usage at cost — a workshop for building your own agents, not a team for hire.

Buy it if

You want to build and own several small AI tools yourself, mix cheap and frontier models per step, and pay provider cost for the intelligence.

Skip it if

You want finished agents or a predictable bundled bill — Gumloop or Relevance package the judgement work for you.

What MindStudio does — the jobs it finishes

Workflows are built only from verified capabilities in our listing research — no vendor copy, nothing invented.

A lead-research agent you own

Starts You feed a list of companies or a new lead arrives

  1. The agent gathers public information with a low-cost model
  2. A stronger model scores fit against your criteria and drafts the summary

You get A scored, summarised lead list in your sheet or CRM

Research tooling without a per-seat research subscription

Document processing as an internal app

Starts Contracts, applications or forms arrive

  1. The agent extracts the fields you defined
  2. Results deploy as a small web app your VA or clients can use

You get Structured records from unstructured documents

A bespoke internal tool instead of a generic SaaS tier

An agent exposed as an MCP server

Starts Another assistant or workflow needs your agent's skill

  1. The agent is published as an MCP server
  2. Claude, ChatGPT or your automations call it as a tool

You get Your custom logic callable from the tools you already use

Build once, reuse everywhere your stack reaches

Where it stops

  • Not a library of finished agents — you design and maintain each one
  • Platform fee excludes model usage on every plan, including Free
  • Business governance (SSO, audit logs, self-hosting) is custom-priced

Getting started

Before you start

  • A clear job for the first agent — one agent only on Free
  • A small prepaid usage balance for model calls

The ongoing admin

  • Top up and watch the prepaid usage balance
  • Review agents when a provider changes a model's price or behaviour
Step 1

Build the first agent on Free

One agent and 1,000 runs a month is enough to validate the editor and the deployment surface you need.

Step 2

Choose models per step deliberately

Cheap models for extraction, frontier models only where judgement matters — the at-cost router makes this the main cost lever.

Step 3

Set a budget before you scale runs

Usage draws on a prepaid balance on every plan; workspace budgets and alerts turn a variable bill into a capped one.

Time to first useful result: An afternoon for a first working agent; days to feel fluent in the editor · no-code 4/5

Full pricing — checked 2026-11-06

Platform fee plus at-cost usage: a Free plan ($0, 1 agent, 1,000 runs/month), Individual at $20/month billed monthly or $16/month billed annually with unlimited agents and runs, and Business custom. On every plan, AI model and service usage is billed separately at provider cost with no markup, drawn from a prepaid balance.

Free

$0/month + usage — 1 agent, 1,000 runs/month

Flat / month · $0/mo for one person

  • Visual agent builder
  • Access to 200+ models via the service router or your own keys

Allowance 1 agent; 1,000 runs/month

Limit 1 agent

Limit 1,000 runs/month

Limit Usage billed at cost

Solo pick

Individual

$16/month billed annually ($20 billed monthly) + usage — unlimited agents and runs

Flat / month · $16/mo for one person

  • Unlimited agents and runs
  • 200+ models at cost, no markup
  • Slack community

Allowance Unlimited agents and runs

Limit Single user

Limit Usage billed separately at provider cost

Business

Custom — team workspace and governance

Flat / month · $null/mo for one person

  • Granular permissions, SSO and audit logs
  • Budgets and usage alerts
  • Self-hosting and custom domains

Limit Sales-led pricing

PlanPublished priceSolo monthlyBilled byLimits & overage
Free$0/month + usage — 1 agent, 1,000 runs/month$0/moFlat / month1 agent · 1,000 runs/month · Usage billed at cost
IndividualSolo pick$16/month billed annually ($20 billed monthly) + usage — unlimited agents and runs$16/moFlat / monthSingle user · Usage billed separately at provider cost
BusinessCustom — team workspace and governance$null/moFlat / monthSales-led pricing
Mandatory add-onModel and service usage — At provider cost with no markup, drawn from a prepaid balance; per-model token rates published by the vendor. The real variable cost on every plan, including Free

What could make the bill jump?

Usage, not seats: an agent looping on a frontier model burns the prepaid balance; workspace and per-user budgets exist precisely to cap this, and enabling them is the control.

Billing unitper month plus at-cost model usage
Free tier / trialYes — Free plan with 1 agent and 1,000 runs/month; model usage is charged at cost on top.
Pricing clarityUsage-based but predictable
SourceOfficial pricing page ↗
Checked 2026-11-06

Price record · 2026-11-06: Checked 2026-11-06: the vendor's pricing page does not extract as text in this run's fetch, so the figures stand as corroborated by multiple dated reads of the same official page (August 2026, October 2026 and 6 November 2026 reads agree: Free $0 with 1 agent and 1,000 runs/month; Individual $20/month billed monthly or $16/month billed annually with unlimited agents and runs; Business custom; usage passed through at provider cost with no markup). Disclosure: this listing uses that corroborated-dated-reads pattern — the same treatment this directory applies where a rendered page's figures do not extract — and the buyer FAQ repeats it. Some older third-party reviews list retired Starter/Pro/Agency tiers ($16/$48/$140 in a different structure); the three-plan structure above is the one the recent dated reads of the official page agree on. As of November 2026 the entry paid tier is $16/month billed annually plus usage.

At a glance — the standard checklist

The same fields on every deep listing in this job, so agents compare field by field. “Not verified” means our research has not confirmed it yet — it never means “doesn't have it.”

Best for

  • Founders who want to build and own their AI tools
  • Builders mixing cheap and frontier models step by step
  • Anyone who would otherwise stack three niche AI SaaS subscriptions

Not for

  • Founders who want finished agents without building
  • Anyone who needs the platform fee to be the whole bill
  • Teams needing SSO and audit logs without a custom quote

Our assessment of MindStudio

As of November 2026, MindStudio's Individual plan is $16/month billed annually plus model usage at cost — and that split is the product's whole proposition. Where Gumloop or Relevance sell you finished agent capacity, MindStudio sells you the workshop: a visual block editor where a solo founder assembles agents from model calls, data steps, logic and integrations, then deploys each one as a web app, a browser extension, an automation, a webhook or an MCP server other tools can call. Access to more than 200 models through a built-in router — or your own API keys — means the same agent can use a cheap model for extraction and a frontier model for the one step that needs judgement, with usage passed through at provider cost and no platform markup. Founders use it to build internal tools a SaaS subscription would otherwise cover: lead-research agents, document processors, content pipelines, support-draft assistants and small client-facing apps. The trade-offs are builder-shaped: you design, test and maintain the agent yourself, the Free plan allows only one agent and 1,000 runs a month, and the platform fee is never the whole bill — usage is metered separately on every tier, controlled by prepaid balances and budgets rather than by a bundled allowance.

Pros — our take

  • At-cost model access with no markup across 200+ models
  • One builder, many deployment surfaces — apps, automations, MCP servers
  • Unlimited agents and runs on a $16/month billed annually platform fee

Cons — our take

  • Usage is always extra — the headline fee is never the whole bill
  • You build and maintain the agents; there is no done-for-you team
  • Free plan's single agent is a learning tier, not a working one
Solo Fit input · no-code4/5 — can a non-technical owner run it?
Solo Fit input · setupAn afternoon for a first working agent; days to feel fluent in the editor to the first useful result
Solo Fit input · human controlActs + logs everything
Solo Fit input · pricing clarityUsage-based but predictable

Solo Fit stays the public score: the qualitative roll-up of these four published inputs under the How We Rank method. At-cost model access and unlimited agents carry it; build-it-yourself setup is the trade.

Three to compare before you decide

Gumloop$37/mo · Solo-Ready

The difference: Gumloop sells finished agent capacity at a flat $37 with credits bundled; MindStudio sells the builder and passes model costs through at cost.

Make$9/mo · Solo-Ready

The difference: Make automates between apps on a canvas with bundled credits; MindStudio builds standalone AI apps and agents you deploy yourself.

n8nSee pricing · Technical

The difference: n8n is the open-source, self-hostable workflow engine; MindStudio is hosted, AI-first and deploys agents as apps and MCP servers.

Questions buyers ask about MindStudio

What does MindStudio cost for one person as of November 2026?

Free is $0 for 1 agent and 1,000 runs a month. Individual is $20/month billed monthly or $16/month billed annually for unlimited agents and runs. On every plan, model and service usage is billed separately at provider cost with no markup, from a prepaid balance — so the true bill is the platform fee plus your agents' usage.

How were these prices verified?

Disclosure: the vendor's pricing page did not extract as text in this run's fetch, so the figures are corroborated by multiple dated reads of the same official page (August 2026, October 2026 and 6 November 2026) that agree on Free $0, Individual $20/$16 and Business custom, with at-cost usage. Check the live page before budgeting, as with any corroborated listing in this directory.

What does 'usage at cost' actually mean?

MindStudio passes provider model costs through with no markup — language models per million tokens at the provider's own input/output rates, image generation per image, transcription per minute — drawn from a prepaid balance you can cap with workspace budgets and auto-recharge rules.

Can I use my own API keys?

Yes — the service router needs no keys, and bringing your own keys is supported on the plans the dated reads describe; either way the underlying model cost is yours, at the provider's rate.

Stacks & next steps

MindStudio is not in a published stack yet — stacks are edited buying guides, and memberships are added only when the combination has been worked through.

Compare all run operations & workflows agents → · How we rank → · Get a weekly shortlist →

Provenance — where this page’s facts came from