# Plans and limits

Your organization's plan sets what it pays and its limits: how many workspaces and people it can
have, how much AI work it can do each month, and more. This page explains how a plan is priced,
lists every limit a plan can set, says what happens when you reach one, and where to see your
organization's own plan and usage.

<Availability permission="The Owner and Admin roles see the plan and its usage, under Billing & Payments. Anyone who can open the Features or Knowledge base list sees its counter." />

## Key concepts

**Plan.** What your organization subscribes to. Its limits apply across all your workspaces.

**Allowance.** What your plan includes of something each billing period, such as the AI credits
it includes each month, its **Included AI credits**. It starts again from zero at the start of each
billing period: what you don't use doesn't roll over. Past it, extra use is billed at your plan's
rate while overage is on, or work stops until the next billing period while overage is off.

**AI credit limit.** The most AI credits your organization can use in a day, a week or a billing
period: the **Daily AI credit limit**, the **Weekly AI credit limit** and the **Monthly AI credit
limit**. Productmind sets them in each plan, and a plan without one has no such limit. A limit
counts the AI credits you pay for too, and it is never billed: past it, work stops until it
resets, whatever the overage switch says and whatever your Pay-as-you-go balance holds.

**Billing period.** Your month. It counts from your subscription date, not from the calendar:
it starts on the day and at the time of day your subscription started, and ends at the same moment
a month later. On the plan with no monthly fee, it counts from the start of the day your
organization was created. Allowances are monthly even when you pay for a year ahead.

**Week and day.** The weekly and daily limits count from the start of your subscription, not from
your billing period. A week is always 7 days, and it rolls on across billing periods: a week can
start in one billing period and end in the next, and no week is shorter. A day starts at your
subscription's time of day, which isn't necessarily midnight. On the plan with no monthly fee, both
count from the start of the day your organization was created. The **Usage** tab and the sidebar
say when each one resets.

**AI credits.** The unit AI work is measured in. All AI work draws on the same AI credits:
Productmind's agents, and coding sessions and scans in Productmind's cloud. Every model's
tokens use AI credits in proportion to what they cost. A coding session that runs in Productmind's
cloud also uses 500 AI credits for each web search its agent makes. Coding sessions on your own
machine use none of your AI credits. See [Usage](/user-guide/settings/billing-and-payments/usage).

**User-hosted session.** A coding session the Productmind agent runs on your own machine, on your
own Claude Code or Codex account. If your plan charges for them, each one is charged once, at your
plan's per-session rate, the first time it runs, and never again for that session.

**Cap.** A limit on how many of something you can have, such as workspaces, people or a
workspace's features, or on how many referrals each person can send. A cap counts across your
whole organization, in each workspace, or for each person, as the tables below say.

**Unlimited.** A plan can leave a limit out, and then there is no limit.

**Not included.** A plan can set a limit to zero. The feature is then not part of the plan, and
Productmind says so when you try to use it.

**Pay as you go.** A plan can include none of an allowance and charge every unit at its rate
instead, from the first one, paid from your Pay-as-you-go balance.

**Pay-as-you-go balance.** A prepaid balance, in euros, that an Owner or Admin can top up. Pay as
you go usage is paid from it, and for some allowances, usage beyond the plan can be drawn from it
instead of stopping. It is money, not usage: AI credits measure your AI work, and the balance pays
for it.

## How a plan is priced

A plan can have a monthly price, per user or for your whole organization, with 20% off when you
pay for a year ahead. A plan with no monthly fee reads **Pay-as-you-go**. Everything you use is
priced on top, per unit:

| Price | What it is |
| --- | --- |
| **Price per 1M AI credits** | What your plan charges for AI work. Every AI task, from a chat message to a coding session or a scan in Productmind's cloud, uses AI credits, all at this one price. |
| **Typical tasks** | What a chat message, a background draft, a coding session in Productmind's cloud, a feature deep read and the scan of one repository usually cost on the plan. They are measured medians: each task costs the AI credits its own work uses. |
| **Unit prices** | Per session on your own machine, per hour of each sandbox size, per minute of voice interviews and of the notetaker, and per hour of a review by a Productmind Architect. |

You see these prices when you choose a plan during setup, on the
[Plan](/user-guide/settings/billing-and-payments/subscription-plan) tab's **Change plan**, in the dialog
that offers a larger plan when you reach a limit, and on the pricing page of the Productmind
website. This guide doesn't list them, because they change from time to time.

Feature discovery and application scanning are on every plan, and their AI work is charged in AI
credits like any other.

## The limits a plan can set

**Billing & Payments › Usage** shows them in these groups, with the same names.

### Software

| Limit | What it counts | Counted |
| --- | --- | --- |
| **Max Workspaces** | The workspaces in your organization. | Across your organization |
| **Users** | The people in your organization. Pending invitations count too. On a per-user plan, the bill counts active members only: see [Member limit and billed members](/user-guide/settings/identity-and-access/users#member-limit-and-billed-members). | Across your organization |
| **Studies per workspace** | The researches on a workspace's [Agentic interviews](/user-guide/feedback/agentic-interviews) page. | In each workspace |
| **Features per workspace** | The features in a workspace. Features copied from a template when you start an application from one don't count. | In each workspace |
| **Knowledge base sources** | The sources in a workspace's knowledge base. The documents Productmind creates for every workspace don't count. | In each workspace |
| **Personal referrals** | The referrals each person can send to bring other organizations to Productmind. | For each person |

### AI

Every AI limit counts across your whole organization.

| Limit | What it counts | Pay-as-you-go |
| --- | --- | --- |
| **Included AI credits** | The AI credits your plan includes each billing period for all your AI work: Productmind's agents (chat, specifications, the knowledge base, interviews, the feedback widget and generated content), and coding sessions and scans that run in Productmind's cloud. Agents and coding share them, each model at its own rate. Past them, extra use is billed at your plan's price per 1M AI credits while overage is on. | Yes |
| **Monthly AI credit limit** | The most AI credits all your AI work can use in a billing period, the ones you pay for included. A safeguard: past it, work stops until the next billing period starts. | No |
| **Weekly AI credit limit** | The same, per week of 7 days, the ones you pay for included. A safeguard against runaway use. | No |
| **Daily AI credit limit** | The same, per day, the ones you pay for included. A safeguard against runaway use. | No |
| **Voice interview minutes** | Minutes of voice interviews, where the AI interviewer talks to a customer, each billing period. | Yes |
| **AI notetaker minutes** | Minutes of customer meetings and internal meetings that Productmind's notetaker attends, from when it joins the meeting until it leaves, each billing period. | Yes |
| **Included cloud allowance (EUR)** | An amount in euros your plan includes each billing period for time in Productmind's cloud. Live previews and coding sessions on a sandbox of any size draw on it by the second, each at its size's price per hour. Past it, each size is billed at its own price while overage is on. Runs on your own machine are free. | Yes |
| **Sandbox hours limit, Small** | The most hours live previews and coding sessions can run in Productmind's cloud on a **Small** sandbox (2 vCPU, 8 GB memory, 50 GB storage) each billing period. A ceiling: past it, that size stops until the next billing period. Their time still draws on the **Included cloud allowance**. | No |
| **Sandbox hours limit, Medium** | The same on a **Medium** sandbox (4 vCPU, 16 GB memory, 60 GB storage), the default size. | No |
| **Sandbox hours limit, Large** | The same on a **Large** sandbox (8 vCPU, 32 GB memory, 100 GB storage). | No |
| **User-hosted sessions** | Coding sessions the Productmind agent runs on your own machine each billing period, including scans. Each session counts once, the first time it runs: its plan, its build, its scan, or its first change in a Live session. Its AI runs on your own account and uses no AI credits, and its previews are free. | Yes |
| **Sessions claimed over MCP** | Sessions your own coding assistant claims over MCP each billing period, each counted once. Never charged. | No |

**Pay-as-you-go** says whether the allowance can run into overage, paid from your Pay-as-you-go
balance. That happens only when your plan has a pay-as-you-go rate for it and its overage switch
on the **Usage** tab is on. See [Usage](/user-guide/settings/billing-and-payments/usage).
The same allowances can also be pay as you go from the first unit, when your plan includes none
of them and sets a rate: the plan comparison then reads **None, pay as you go**. The AI credit
limits and the sandbox hours limits never run into overage: they are ceilings, not allowances, so
neither overage nor your Pay-as-you-go balance lifts them.

**A user-hosted session is charged once.** When your plan charges for sessions on your own
machine, the charge is your plan's per-session rate, taken the first time the session runs there.
Later steps of the same session (answering its questions, building its plan, requesting changes
to its pull request, more changes in a Live session) are never charged again. A session that
starts in Productmind's cloud and moves to your machine is charged once, when it first runs there,
on top of the AI credits it used in the cloud.

**Time is billed by the second.** Voice interview minutes, notetaker minutes and sandbox time are measured in seconds. What pay-as-you-go charges is the seconds you used at
your plan's rate per minute or per hour, rounded to a whole cent on the billing period's total, so
a 66-second call is never charged as two minutes and a 6-minute preview is never charged as an
hour. The **Included cloud allowance** is drawn by the second too: a 6-minute preview on a sandbox
priced at €0.50 an hour uses €0.05 of it. The **Usage** tab shows minutes rounded up to the next
whole minute.

**Each sandbox size has its own price, and every size shares one allowance.** A live preview or a
coding session in Productmind's cloud draws on the **Included cloud allowance** at its sandbox
size's price per hour. With prices of €0.25, €0.50 and €1.00 an hour for Small, Medium and Large, a
€20.00 allowance covers about 80 hours of Small, 40 of Medium or 20 of Large, or any mix of them.
Once it is used up, one overage switch decides for every size: with overage on, each size is charged
at its own price per hour from your Pay-as-you-go balance, and your invoices still show each size
on its own line; with overage off, new live previews and coding sessions in the cloud are refused
until the next billing period. When the allowance is **Unlimited**, sandbox time costs nothing and
only the sandbox hours limits apply; when your plan includes none of it (**None, pay as you go**),
every second is charged from the first one. A **Sandbox hours limit** caps one size whatever the
allowance and whatever overage does, and at zero that size isn't included. A size your plan sets no
price for can't run in Productmind's cloud while your plan has an allowance. Scans don't run in a
sandbox and use none of it. See [Sandboxes](/user-guide/engineering/sandboxes).

AI credits are cost-weighted: every model's tokens use AI credits in proportion to what they
cost, so work done by a more capable, more expensive AI model uses more of your allowance than the
same work done by a lighter one. Each web search the agent of a coding session in Productmind's
cloud makes uses 500 AI credits on top of its tokens. Agents and coding sessions draw on the same
AI credits, and the **Usage** tab shows how many of them each used this billing period.

**How included AI credits and the limits work together.** Take a plan with 10M **Included AI
credits** and a **Weekly AI credit limit** of 5M. Two heavy weeks use the whole 10M. From then
on, with overage on, you keep working and pay for each AI credit from your Pay-as-you-go balance,
but still at most 5M a week. With overage off, work stops until the next billing period. A limit
can also stop you before the allowance runs out: with a weekly limit of 1M, you can use at most
about 4M to 5M of a 10M allowance in a billing period, and the rest expires when the period ends.

The plan comparison, under **Change plan** on the **Plan** tab, also lists what else each plan
includes, such as its support and its uptime SLA.

**Single sign-on** depends on your plan. See
[Single sign-on](/user-guide/settings/identity-and-access/single-sign-on) and
[Features that depend on your plan](#features-that-depend-on-your-plan).

**Sandbox sizes.** You can pick any of the three sandbox sizes, **Small**, **Medium** and
**Large**, on every plan. Your plan sets what an hour of each costs and the **Included cloud
allowance** they all draw on, and it can cap a size, or leave it out, with its sandbox hours
limit. See
[Sandboxes](/user-guide/engineering/sandboxes).

## Seeing your plan and usage

### To see your plan's limits and what you've used

1. In the sidebar, choose **Billing & Payments**.
2. Choose **Usage**.

On the **This billing period** card, the **AI** and **Cloud** sections show each allowance: what each use
drew, their **Total usage**, the **Allowance** (or that it is **Unlimited** or **Not
included**), the **Balance** between the two, and a bar of how much of the allowance you have
used. One your plan includes none of, which you pay as you go, shows its uses and one **Total
usage** line, what they were charged. A daily, weekly or monthly AI credit
limit, or a sandbox hours limit, is a line under its allowance with what you have used this day,
week or billing period (the tab shows its dates).

Your plan's other limits are on **Plan**, in your plan's column. Each shows
**Unlimited** when your plan sets no limit, and otherwise:
- a limit that counts across your organization shows a bar with how many you have now, such as
  **used 3 of 10** for **Max Workspaces** and **Users**;
- a limit that counts in each workspace or for each person shows only its value, such as **12
  included**. Features, knowledge base sources and referrals show their count elsewhere, as below.
  **Studies per workspace** has no counter.

To compare plans, choose **Plan**, then **Change plan**. See
[Billing & Payments](/user-guide/settings/billing-and-payments/overview).

### Counters on the lists

When your plan caps them, these lists show a counter such as **8 / 10**:
- **Users**, on [Identity & Access](/user-guide/settings/identity-and-access/users), under the
  toolbar;
- **Features**, in each workspace, beside the number of records. It counts what the cap counts, so
  it leaves out features copied from a template and can read lower than the list. See
  [Features](/user-guide/product/features);
- **Knowledge base**, in each workspace, beside the number of records. See
  [Knowledge base](/user-guide/knowledge-base).

A counter turns red at 90% of its cap. Once the **Features** counter reaches its cap, choose it to
see what your plan allows and how to raise it. A list with no cap on your plan shows no counter.

Your own referrals are counted on the **Refer Productmind to a friend** card at the bottom of the
sidebar, for example **12 of 12 referrals left**, and as **Sent referrals** on the **Refer
Productmind to a friend** page that its **View all** link opens. See
[Finding your way around](/user-guide/get-started/finding-your-way-around#the-bottom-of-the-sidebar).

## Features that depend on your plan

Besides its limits, a plan decides which of these features your organization can use. The
**Plan** tab shows which of them your plan includes, and its **Change plan** which plans include
each one, and any other part of
Productmind a plan leaves out, which then works the way described below.

| Feature | What it covers |
| --- | --- |
| **Lineage** | The **Lineage** page, and the lineage of each feature, feature request, task, session, release and customer conversation. See [Lineage](/user-guide/product/lineage). |
| **Decision records** | Recording decisions on features and feature requests, their **Decisions** tabs, and the decision rules on **Settings > Decision records**. See [Decision records](/user-guide/settings/workspace-settings/decision-records). |
| **Audit log** | Reading and exporting the audit log, and choosing how long it's kept. Productmind records changes on every plan. See [Audit log](/user-guide/settings/identity-and-access/audit-log). |
| **Agent customization** | [Automations](/user-guide/engineering/automations), agent [Definitions](/user-guide/engineering/agents/definitions), and your workspace's own [Model classes](/user-guide/engineering/agents/model-classes). |
| **Support desk import** | Bringing conversations in from Freshdesk or Zendesk, by hand or automatically, and the **Support desk** settings. See [Freshdesk and Zendesk](/user-guide/settings/freshdesk-and-zendesk). |
| **Single sign-on** | Signing in through your own identity provider. See [Single sign-on](/user-guide/settings/identity-and-access/single-sign-on). |
| **Bring your own model key** | Running agents and coding sessions on your organization's own Anthropic, OpenAI or LiteLLM account. See [Model provider](/user-guide/engineering/agents/model-provider). |
| **Workspaces** | Creating your organization's workspaces, and managing each one in [Workspace settings](/user-guide/settings/workspace-settings/workspace): renaming it, its **Permissions** tab, resetting and deleting it. Every plan includes it, and **Max Workspaces** sets how many workspaces it allows. See [Organizations and workspaces](/user-guide/get-started/organizations-and-workspaces#creating-a-workspace). |
| **Marketing** | The **Marketing** page, where you create Marketing releases and turn the releases, features, feature requests and customer feedback they announce into public release notes, blog posts and customer emails. Scale and Enterprise include it. See [Marketing](/user-guide/product/marketing). |

### When your plan doesn't include a feature

- **Its menu entry, tab or button shows a lock.** It stays where it is, so you can still find it.
- **Opening it shows a page, or a dialog, naming the plan that includes the feature**, or the
  Enterprise plan when none of the self-service plans does. An Owner or Admin sees **View plans**,
  which opens the **Plan** tab. Anyone else is asked to ask an Owner or Admin to
  upgrade.
- **Automations are paused, not deleted.** When a plan no longer includes agent customization,
  your automations stop starting new runs, keep all their settings, and run again once your plan
  includes them.
- **Customized model classes fall back to Productmind's.** When a plan no longer includes agent
  customization, your workspace's own model lists are kept but not used: coding sessions and AI
  agents run on Productmind's lists until your plan includes it again.
- **Decision prompts switch off.** When a plan no longer includes decision records, Productmind
  stops asking for, and requiring, a decision when a status changes, and doesn't record decisions
  sent with a status change. Your decision rules are kept and apply again once your plan includes
  them.
- **Support conversations stop arriving automatically.** When a plan no longer includes support
  desk import, tickets your helpdesk sends are not imported, and importing by hand is unavailable.
- **Single sign-on stops working, but stays visible.** When a plan no longer includes single
  sign-on, members can no longer sign in through your identity provider. An Owner or Admin can
  still see the connection and remove it.
- **Your model key is suspended, but stays visible.** When a plan no longer includes bring your own
  model key, work set to use your key is refused rather than run on Productmind AI credits, and
  your Owners and Admins are told. An Owner or Admin can still see the key and remove it, which
  moves new work back to AI credits.
- **Your marketing content is kept.** While your plan doesn't include Marketing, your customer
  release notes, blog posts, their published and approved versions and the log of customer emails
  sent are kept, and come back once your plan includes it.

## Reaching a limit

**Caps stop you creating more.** Creating a workspace, adding a knowledge base source, creating
a research or adding a feature beyond your cap is refused. Free a place, for example by deleting a
workspace you no longer use, or move to a plan with a higher cap. Once you have sent all your
personal referrals, you can't send another.

**Features copied from a template don't use up the feature cap.** Starting an application from a
template copies its feature specifications into your workspace whatever your cap is, and none of
them count toward it. Every other feature does, including the ones feature discovery creates,
although feature discovery itself is never stopped by the cap.

**Interview minutes stop new interviews and recordings.** When your voice interview minutes are
used up, a new voice interview can't be set up until the next billing period, unless
pay-as-you-go covers it. The same goes for asking the notetaker to record a meeting when your
AI notetaker minutes are used up.

**Sandbox limits and the cloud allowance stop new previews and coding sessions in the cloud.**
Productmind checks the size's **Sandbox hours limit** first: once it is reached, that size stops
until the next billing period, with "Monthly limit of Medium sandbox hours reached. It resets at
the start of your next month.", even with overage on and money in your balance. Then it checks the
**Included cloud allowance**: while some of it is left, the run starts at no charge. Once it is used
up, the run starts only with overage on and a Pay-as-you-go balance that covers one hour at the
size's price; otherwise Productmind shows "Overage is off for cloud usage and your cloud allowance
is used up. Turn overage on under Billing & Payments > Usage to continue." A preview or a session
that is already running keeps running, and its time past the allowance is charged.

**Daily, weekly and monthly AI credit limits stop the work until they reset.** They apply to
agents and coding sessions alike, because both use the same AI credits, and they count the AI
credits you pay for too. Neither overage nor your Pay-as-you-go balance lifts them. They also stop
coding sessions that are already running in Productmind's cloud, within about a minute: see
**Work that has started** below. On the
**Usage** tab, each one's row says "Limit - blocks until it resets", and when it resets. A day
starts at your subscription's time of day, which isn't necessarily midnight, and a week every 7
days from the start of your subscription, whatever the weekday and whatever the billing period.
The monthly limit resets when your next billing period starts. Productmind says "Daily AI credit
limit reached. Please try again after it resets, or move to a plan with a higher limit.", and the
same for the weekly and monthly limits. When no plan you can switch to raises the limit, the
sentence ends "Please try again after it resets.", and when Productmind can't tell which plans
raise it, "Please try again after it resets, or upgrade your plan."

**Monthly allowances stop the work, unless pay-as-you-go is on.** When a monthly allowance is
used up, new work of that kind is refused until the next billing period. When your plan has no
pay-as-you-go for its **Included AI credits**, Productmind says "Your plan's included AI credits
for this month are used up. Please upgrade your plan.", or, when it includes none and charges
none, "AI credits are not included in your plan. Please upgrade your plan." For allowances that
support it, an Owner or Admin can turn on pay-as-you-go on the **Usage** tab: usage beyond the
allowance is then drawn from your Pay-as-you-go balance, as long as it covers it. While it is off,
Productmind shows, for example, "Overage is off for AI credits and your allowance is used up. Turn
overage on under Billing & Payments > Usage to continue.", with an **Open Usage** action. See
[Usage](/user-guide/settings/billing-and-payments/usage). Pay-as-you-go never goes past a limit,
though: once your plan's **Monthly AI credit limit** is reached, work stops until the next billing
period, even with overage on and money in your balance.

**When several things stop the work at once,** Productmind names the one that waiting won't fix:
the monthly limit first, then your used-up **Included AI credits** (with overage off, with no
pay-as-you-go on your plan, or with a balance too low to pay), then the week, then the day. So you
are never told to wait for your week to reset when your included AI credits are used up as well.

**Pay-as-you-go allowances stop the work only when spending is paused or can't be paid.** On an
allowance your plan includes none of, every unit is charged. If an Owner or Admin has paused its
spending on the **Usage** tab, Productmind shows, for example, "Pay-as-you-go spending is paused
for AI credits. Turn it back on under Billing & Payments > Usage to continue.", with an **Open
Usage** action. Pausing AI credits stops agents and coding sessions alike. Pausing the cloud
allowance stops new live previews and coding sessions in the cloud of every size, with
"Pay-as-you-go spending is paused for cloud usage. Turn it back on under Billing & Payments > Usage
to continue." If the balance can't cover the next piece of work, it asks you to top up.

**On the plan with no monthly fee, some work needs pay-as-you-go activated.** Until your
organization makes its first top-up, which also saves its payment method, pay-as-you-go covers only
a small amount of AI use by Productmind's agents, such as chat. It doesn't cover coding sessions
and scans that run in Productmind's cloud, even though they use the same AI credits. They,
sessions on your own machine, live previews and interviews that would be paid from your
Pay-as-you-go balance are refused with the message "Pay-as-you-go is not activated", whose
**Activate** opens the first top-up with EUR 10 already entered. See
[Before you activate pay-as-you-go](/user-guide/settings/billing-and-payments/activating-pay-as-you-go#before-you-activate-pay-as-you-go).

**Work that has started isn't cut off, with two exceptions.** Limits are checked before work
starts: an AI reply, a coding session's next step, a live preview. A live preview that is already
running keeps running, and so does an AI reply that has started. A coding session running in
Productmind's cloud is stopped:

- if its running step takes your Pay-as-you-go balance below zero by more than the cost of one
  typical coding step. Top up, then send a message in the session to continue;
- within about a minute of your organization reaching its daily, weekly or monthly AI credit limit.
  Every coding session your organization is running in Productmind's cloud stops then, whatever
  work reached the limit. The session says "Stopped: the organization reached its AI credit
  limit. Send a message to continue once it resets.", which limit stopped it and when it resets,
  and **See plans** opens **Usage limit reached**. Once the limit has reset, send a message in the
  session to continue; a message sent before then is refused with the limit's message.

Either way, a build saves its work in progress to its branch first, and its pull request opens when
the build continues. Coding sessions on your own machine use none of your AI credits, so neither
stop applies to them. See
[Usage](/user-guide/settings/billing-and-payments/usage#how-usage-behaves).

**A scan starts only when your balance covers it.** Before a scan of your repositories starts,
Productmind estimates what it will cost, as a range from a typical scan to a larger one. On a plan
that charges scans from your Pay-as-you-go balance, the scan starts only when the balance covers the
lower end of that range; otherwise it is refused with the same message as other work, such as the
top-up message. A scan that runs on your own machine uses no AI credits, so the estimate doesn't
apply to it. See [Applications](/user-guide/engineering/applications#what-a-scan-costs).

**A scan that runs out pauses.** If your balance runs out between the steps of a scan, the scan
pauses instead of failing: it shows **Paused: top up to continue**, and continues by itself after a
top-up, or when you choose **Resume**.

**When you reach a limit your plan has no pay-as-you-go for**, such as a cap, a daily, weekly or
monthly AI credit limit, or a monthly allowance without a pay-as-you-go rate, Productmind shows **Usage limit reached**,
and the dialog's message names the limit. For a daily, weekly or monthly
limit, the dialog then says when the limit resets, in your own time zone, for example
"Your daily limit resets Fri 02:00.", or, for a reset more than six days away,
"Your monthly limit resets Oct 29, 08:00.": the time the limit's row on the **Usage** tab shows.
What else it shows depends on the limit and the plans:

| When | The dialog shows |
| --- | --- |
| A plan you can switch to raises the AI credit limit, and you are an Owner or Admin | Only the plans that raise it, each with its price and its price per 1M AI credits. **Open Usage** opens **Billing & Payments › Usage**, and **View plans** the **Plan** tab. **Not now** closes the dialog. |
| A plan raises the AI credit limit, but you can't change the plan | "Ask an Owner or Admin to move to a plan with a higher limit.", and **Got it**. |
| No plan you can switch to raises the AI credit limit | "No plan you can switch to raises this limit.", **Got it** and **Contact us**. |
| Any other limit, or Productmind can't tell which plans raise the AI credit limit | Each plan's price and its price per 1M AI credits, **Not now** and **View plans**. |

No plan raises a limit when every plan you can switch to sets the same one, or when your plan is
agreed with Productmind or you pay by invoice. **Contact us** then opens a new support request in
the [Support center](/user-guide/help-and-support): a question titled, for example,
"Raise our daily AI credit limit", with what your organization used of the limit, your plan and
when the limit resets already filled in. Check it, and send it. If your role can't open support
requests, **Contact us** starts an email to Productmind support instead.

**Coding sessions on your own machine use no AI credits.** A session built by the Productmind
agent on your computer uses your own coding agent's account, and its live preview runs on your
machine, so neither uses your plan's AI credits or its cloud allowance. If your plan
charges for sessions on your own machine, each one is charged once, at your plan's per-session rate, the
first time it runs, and counts toward **User-hosted sessions**. Previews on your machine stay free.
Sessions your own coding assistant claims over MCP are counted, never charged. See
[Productmind CLI](/user-guide/tools/productmind-cli).

## Troubleshooting

### "Usage limit reached"

The dialog's message names the limit you reached, and your usage is on **Billing & Payments ›
Usage**, which an Owner or Admin can check. Then wait for the allowance or the limit to start again (for a daily, weekly or monthly limit, the dialog says when), or move
to a plan with a higher limit. When no plan you can switch to raises the limit, the dialog says so:
choose **Contact us** to ask Productmind to raise it.

### "Your Pay-as-you-go balance is ..., and ... needs at least ... available"

Pay-as-you-go is on, but the balance doesn't cover the next piece of work. The message says what the
balance holds and what the work needs, for example "Your Pay-as-you-go balance is EUR 4.57, and a
coding step needs at least EUR 6.24 available. Top up to continue." An Owner or Admin can top up the
balance on the **Usage** tab, or with **Top up** on the message, which opens the top-up dialog with
the amount that covers it already entered. A scan that paused for it continues after the top-up.

### "Overage is off"

A monthly allowance is used up and an Owner or Admin has left its overage off. They can turn it on
(**Overage on**) on the **Usage** tab, and top up the Pay-as-you-go balance if it is low. See
[To turn overage on or off](/user-guide/settings/billing-and-payments/usage#to-turn-overage-on-or-off).

### "Pay-as-you-go spending is paused"

An Owner or Admin paused spending on an allowance your plan includes none of. They can turn it
back on (**Spending on**) on the **Usage** tab. See
[To pause or resume pay-as-you-go spending](/user-guide/settings/billing-and-payments/usage#to-pause-or-resume-pay-as-you-go-spending).

### "Activate pay-as-you-go to keep going"

Your organization is on the plan with no monthly fee and hasn't activated pay-as-you-go yet. An
Owner or Admin can activate it with one first top-up, which also saves the payment method. See
[To activate pay-as-you-go](/user-guide/settings/billing-and-payments/activating-pay-as-you-go#to-activate-pay-as-you-go).

### "AI credits for coding sessions are not included in your plan"

Your plan doesn't include AI credits for cloud coding sessions. Build the session on your own
machine with the [Productmind CLI](/user-guide/tools/productmind-cli), or move to a plan that
includes them.

### "Daily AI credit limit reached", "Weekly AI credit limit reached" or "Monthly AI credit limit reached"

Your organization has used its AI credits for the day, the week or the billing period, across
agents and coding sessions, the ones you pay for included. The limit resets at the time
**Usage limit reached** says, which the **Usage** tab also shows on its row: the monthly limit when
your next billing period starts. Wait for it, move to a plan with a higher limit, or build coding
sessions on your own machine, which use none of your AI credits. If the message ends at
"Please try again after it resets.", no plan you can switch to raises the limit: choose
**Contact us** in the dialog to ask Productmind to raise it. Neither overage nor your
Pay-as-you-go balance lifts these limits.

A coding session that was running in Productmind's cloud when the limit was reached stops within
about a minute, with "Stopped: the organization reached its AI credit limit. Send a message to
continue once it resets." A build saves its work in progress to its branch first. Choose
**See plans** in the session to open **Usage limit reached**, or wait for the limit to reset and
send a message in the session to continue it.

### "Your plan's included AI credits for this month are used up"

Your organization has used the AI credits your plan includes this billing period, and your plan
has no pay-as-you-go for more. Wait for the next billing period, build coding sessions on your own
machine, or move to a plan that charges for AI credits past its allowance. "AI credits are not
included in your plan" means your plan includes none and charges for none.

### "Medium sandboxes are not included in your plan"

The message names the size of the session's sandbox: your plan's **Sandbox hours limit** for that
size is zero, or your plan sets no price for the size while it has a cloud allowance, so a live
preview or a coding session can't run on it in Productmind's cloud. Choose another size for the
sandbox (see [Sandboxes](/user-guide/engineering/sandboxes#to-change-the-size)), build
the session on your own machine with the [Productmind CLI](/user-guide/tools/productmind-cli), or
move to a plan that includes the size. "Monthly limit of Medium sandbox hours reached"
means that size's **Sandbox hours limit** is reached for this billing period: overage doesn't lift
it, so choose another size or wait for the next billing period. "Sandboxes are not included in your
plan" means your plan includes no cloud allowance and prices no sandbox size, so nothing can run in
Productmind's cloud.

### "Sessions on your own machine are not included in your plan"

Your plan doesn't include sessions the Productmind agent runs on your machine. Build the session
in Productmind's cloud instead, or move to a plan that includes them. "Monthly limit of sessions on
your own machine reached" means your plan's **User-hosted sessions** are used up for this billing
period. The same applies to "Sessions claimed by your coding assistant are not included in your
plan".

### "Your plan does not include ..."

The feature you opened, or the action you tried, is not part of your organization's plan. The
page or dialog names the plan that includes it. An Owner or Admin can change the plan on
**Billing & Payments > Plan**, or contact Productmind support for the Enterprise
plan. See [Features that depend on your plan](#features-that-depend-on-your-plan).

## Related topics

- [Billing & Payments](/user-guide/settings/billing-and-payments/overview)
- [Usage](/user-guide/settings/billing-and-payments/usage)
- [Plan](/user-guide/settings/billing-and-payments/subscription-plan)
- [Roles and permissions](/user-guide/reference/roles-and-permissions)
