> For the complete documentation index, see [llms.txt](https://docs.eesel.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.eesel.ai/pricing/overview.md). # Overview What eesel costs: a 7-day free trial with 100 credits, then a monthly plan of credits where a ticket or a chat that wakes your agent is 1 credit and other work uses credits by how much it does. No per You pay for the work your teammate does, as a monthly plan of credits. A ticket or a chat is 1 credit. No per-seat fees, and no limit on how many integrations you connect or agents you run. ## Free trial A new workspace gets **100 credits for 7 days**, with every feature unlocked and no card needed. On the trial your agent can write up to **2 blog posts** and generate up to **20 images**. Both use credits from the 100, like any other work. The trial ends when you've used the 100 credits or the 7 days are up, whichever comes first, and your agents pause. To carry on, choose a plan in **Workspace settings > Billing** (click your agent's name at the top left). Your first month on the plan starts with its full credits. ## What uses credits | Work | Credits | | -------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | A ticket that wakes your agent | 1, however many replies or automations touch it, and 1 even when the run decides not to act | | A conversation in your chat bubble, inline chat, chat link or helpdesk live chat | 1, however many messages go back and forth | | Asking your agent in Slack, Teams, Discord or Google Chat | By how much work it does: a quick answer is 1, a long piece of work more | | A scheduled run, or an automation that isn't about a ticket | By how much work it does, at least 1 | | The browser extension's side panel | On a ticket, it shares that ticket's 1 credit. Anywhere else, by how much work it does | | A phone call | 1 per minute | | An image your agent generates | 1 per image, whatever its size | | A blog post | Several, including 1 for each image in it | | Chatting with your agent in eesel | Free for short chats, like setup questions or a quick ask. A longer chat uses credits for all of its work, by how much it does. A voice chat's minutes count as part of the chat, at 1 credit for each minute started. Images still use 1 credit each | | A voice call in your chat bubble | Part of that conversation's 1 credit. A long call reaches the fair-use limit below, at about 5 minutes, and uses 1 more credit for each further 5 minutes or so | Tickets and chats have fair-use limits that only very long ones reach: a ticket or chat where the agent does an unusual amount of work uses 1 more credit for each extra chunk of it, a chat that picks up again after more than 24 hours idle counts as a new chat, and a ticket still being worked more than 30 days after it started counts again. Most tickets and chats never reach them. Work eesel starts on its own, such as getting set up after you connect an app, never uses credits. **A credit is charged when your agent is woken, not when it acts.** By the time a run decides a ticket is out of scope it has already read the ticket and reasoned about it, so a run that sends no reply and takes no action still uses 1 credit. The way to not pay for a ticket is to stop it waking your agent in the first place, which is what the next section is about. ## Plans Every plan includes every feature and integration. | Credits a month | Price | | --------------- | ------------ | | 500 | $299/month | | 1,000 | $499/month | | 1,500 | $699/month | | 2,000 | $849/month | | 2,500 | $999/month | | 3,000 | $1,149/month | | 3,500 | $1,299/month | | 4,000 | $1,449/month | | 4,500 | $1,599/month | | 5,000 | $1,749/month | | More than 5,000 | Talk to us | Start from your monthly ticket count, since one ticket is 1 credit, and add some for chats, scheduled runs and content. If you guess low, move up a plan any time. For more than 5,000 credits a month, yearly billing or anything custom, ask us in the support chat. Our [pricing page](https://eesel.ai/pricing) has the details. ## Running out and extra credits Your plan's credits reset at the start of each billing month. Unused credits don't roll over. When you've used your plan's credits, your agents keep working on **extra credits at $0.80 each**, up to a **monthly cap on extra spend** that you set. The cap starts at $200. Set it to $0 if you'd rather your agents pause as soon as the plan's credits are used. At the cap, your agents pause until your credits reset. To start them again sooner, raise the cap or move up a plan. Work already in progress finishes, and new work doesn't start. You get an email at 50%, 75% and 100% of the point where your agents pause. The same emails go out on the trial. Extra credits are invoiced each time $100 of them builds up, and the rest at the end of the month. If you often go over, a bigger plan usually costs less. ## Controlling what you use The biggest lever is deciding what reaches your teammate in the first place, and that's done with [automations](/automations-and-schedules/automations-and-schedules.md) rather than billing settings. * **Choose your trigger carefully.** An automation on every customer message picks up more work than one on a narrower event, like a ticket's first message (Zendesk) or a tag being applied (Intercom). * **Describe which events should wake your agent.** The automation's instructions box says which conversations it is for. That part is read before anything runs, so an event it skips starts no run and uses no credit. Leave the box empty and the automation wakes your agent on every event in the mailbox, including tickets you would never act on, and each one uses a credit. Saying "handle everything" in the box is not the same as leaving it empty: it still reads every event, so it still costs something. * **Roll out gradually.** Route one queue, one brand or one topic first. Anything outside it stays with your team and uses no credits. This is also how most teams start: a narrow filter, a look at the results, then widen it. ## Managing your plan Everything lives in your workspace settings. Click your **agent's name at the top left**, open **Workspace settings**, then **Billing**. The page shows your plan, the credits used this month, your usage over time, your cap on extra spend and your invoices. * **Moving up** takes effect straight away. A new billing month starts that day with the new plan's full credits. You're credited for the unused part of your old plan, and extra credits from the old month are billed on the same invoice. * **Moving down** takes effect from your next billing date. You keep your current plan until then. * **Cancelling** takes effect at the end of your billing month. Your agents keep working until then, and you can undo it before that date. No lock-in and no cancellation fee. Billing runs through **Stripe**. Download invoices, update your card or cancel from the same page. ### Digging into what you used **Activity** in the left sidebar lists every task your agent has run, with the credits each one used, or "Free". Filter it by date or by credits used. Or just ask your teammate: it can tell you what it's been working on. See [Reports](/reports/reports.md). ## FAQ
Do I need a credit card to start? No. The free trial needs no payment details. You add a card when you choose a plan.
What happens when I run out of credits? Your agents keep going on extra credits at $0.80 each until they reach your monthly cap on extra spend, then pause until your credits reset. Raise the cap or move up a plan to start them again straight away. On the trial, your agents pause when the 100 credits are used, until you choose a plan.
Do unused credits roll over? No. Your plan's credits reset at the start of each billing month.
Does a long ticket use more credits than a short one? Almost never. A ticket is 1 credit however many replies go back and forth. Only a ticket where the agent does an unusual amount of work, or one still being worked more than 30 days after it started, uses more.
Does a ticket my agent decides to skip use a credit? Yes, 1 credit, if the ticket woke your agent. The credit pays for the run that read the ticket and reached that decision. A ticket that never wakes your agent uses nothing, so narrow what wakes it in the automation's instructions box. See [Controlling what you use](#controlling-what-you-use).
Do I pay for tasks that go wrong? AI resources are used on every task, whether the result is perfect or not, so tasks use credits regardless of outcome. If you're seeing repeated issues or usage that doesn't look right, contact our support team from the in-app chat under **Help and support**. We'll help troubleshoot and make it right.
Can I run multiple agents on one account? Yes. Helpdesk agents, blog writers and e-commerce agents all run under one account, each with its own knowledge and instructions, and they share your plan's credits. There's no charge per agent.
Can I cancel anytime? Yes, from **Workspace settings > Billing**. Your plan runs to the end of its billing month. No lock-in, no cancellation fees.
--- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://docs.eesel.ai/pricing/overview.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.