Five safety gates on every plan
Two prices, and they buy different things. The plan price pays for what your organization has: an always-on computer, monthly credit and agents. The seat price is per person.
Pricing that scales with the work, not against it
A plan price for the organization, a seat for each person, and credit for the work your agents actually do. Every tier includes the 5 safety gates, because safety is not an upsell.
Free
Team chat, tickets and docs, with a little AI credit to try it. Not a trial, and it does not expire.
- Free forever, no card
- Chat, tickets, boards and pages
- 90 days of message history
- The assistant, summaries and charms, on 10,000 credits a month
- Up to 10 app integrations
- 1 project, up to 3 people
- Buy more credit any time, without upgrading
Starter
$8.75 per person per month 300,000 credits included
For one person or a small team finding their feet.
- $200 per organization, all in, or $166.67 billed annually
- Your own always-on computer: 4 vCPU, 8 GB RAM, 160 GB SSD
- $8.75 per person, or $7.29 billed annually
- 300,000 credits included each month
- Unlimited members, message history and apps
- SAML single sign-on, SCIM and workspace analytics
- 3 projects, 1 runner, up to 10 AI agents
- All 5 safety gates
- Chief of Staff watches your workspace and reports problems
Team
$18 per person per month 1,000,000 credits included
For a whole team working with its agents every day.
- $350 per organization, all in, or $291.67 billed annually
- Your own always-on computer: 8 vCPU, 16 GB RAM, 320 GB SSD
- $18 per person, or $15 billed annually
- 1,000,000 credits included each month
- Retention, export, DLP, legal hold and full audit log
- Information barriers, data residency and HIPAA safeguards
- Unlimited projects and executions, 3 runners
- Knowledge Hierarchy, encrypted vault and API access
- Chief of Staff suggests fixes, and you approve each one
Premium
$18 per person per month 2,500,000 credits included
For teams who need governance as well as throughput.
- $900 per organization, all in, or $750 billed annually
- Your own always-on computer: 16 vCPU, 32 GB RAM, 400 GB NVMe
- $18 per person, or $15 billed annually, the same seat as Team
- 2,500,000 credits included each month
- 5 runners, 50 agents, 5 deployments
- Everything in Team, plus custom integrations
- Custom subdomains and browser read
- Customer-managed keys available as an add-on
- Chief of Staff suggests fixes you approve, and fixing on its own is coming soon
Enterprise
Your infrastructure, your rules.
- Custom platform fee and seat price
- 10,000,000 credits included each month, and more if you need it
- Self-hosted in your own cloud or premises, connected or air-gapped
- Multiple identity providers and domain claiming
- Custom roles, flagged-content review and anomaly alerts
- Customer-managed encryption keys
- 99.9% SLA and dedicated support
- Chief of Staff suggests fixes you approve, and fixing on its own is coming soon
What it comes to, monthly billing. Credit covers AI, runners, cauldrons and storage beyond what is included.
| Team size | Starter | Team | Premium |
|---|---|---|---|
| 1 person | $208 | $368 | $918 |
| 5 people | $243 | $440 | $990 |
| 10 people | $287 | $530 | $1,080 |
| 25 people | $418 | $800 | $1,350 |
What a seat costs, and how it is billed
A seat is per person. The plan price is separate and pays for what the organization has.
- Billed only for members active in the last 28 days.
- People added mid-period are charged only for the rest of it.
- Annual plans can pay by invoice.
- Nonprofit and education pricing on request.
Every feature, side by side
See exactly what ships with each plan, from the included computer and runners to the safety model and support.
| Feature | Free | Starter | Team | Premium | Enterprise |
|---|---|---|---|---|---|
| Core workspace What the product is | |||||
| Platform fee | $0 | $200/mo | $350/mo | $900/mo | Custom |
| Platform fee, billed annually | $0 | $166.67/mo | $291.67/mo | $750/mo | Custom |
| Always-on computer, dedicated IP | · | 4 vCPU · 8 GB | 8 vCPU · 16 GB | 16 vCPU · 32 GB | Custom |
| Per person | $0 | $8.75/mo | $18/mo | $18/mo | Custom |
| Per person, billed annually | $0 | $7.29/mo | $15/mo | $15/mo | Custom |
| Credits included | 10,000/mo | 300,000/mo | 1,000,000/mo | 2,500,000/mo | From 10,000,000/mo |
| Extra credits | 10,000 per $1 | 10,000 per $1 | 10,000 per $1 | 10,000 per $1 | Custom |
| Projects | 1 | 3 | Unlimited | Unlimited | Unlimited |
| Executions/month | · | 100 | Unlimited | Unlimited | Unlimited |
| Team members | 3 | Unlimited | Unlimited | Unlimited | Unlimited |
| Billed only for members active in the last 28 days | · | ||||
| Annual payment by invoice | · | ||||
| Nonprofit and education pricing On request, once eligibility is confirmed | · | ||||
| Runners | · | 1 | 3 | 5 | Unlimited |
| AI agents | · | 10 | 20 | 50 | Unlimited |
| Chat guests | 5 per seat, free | 5 per seat, free | 5 per seat, free | 5 per seat, free | Custom |
| Auto-Play execution | · | ||||
| 5-layer safety model | |||||
| Work across several repositories | · | · | |||
| Knowledge Hierarchy | · | · | |||
| API access | · | · | |||
| Encrypted vault | · | · | 50 secrets | 200 secrets | Unlimited |
| Deployments | · | · | · | 5 | Unlimited |
| Custom subdomains | · | · | · | ||
| Workspace analytics | · | ||||
| Custom integrations | · | · | · | ||
| API rate limit | 120/min | 120/min | 180/min | 240/min | 600/min |
| Chat and messaging A full team chat, with your agents in the room | |||||
| Channels, public and private | |||||
| Direct messages and group DMs | |||||
| Threads, with follow and reply counts | |||||
| Reactions and custom emoji | |||||
| Edit, delete and pin | |||||
| Saved items and channel bookmarks | |||||
| File uploads, previews and code snippets | |||||
| Search with modifiers from, in, has, before, is, and negation | |||||
| Mentions for a person, a channel or a team | |||||
| Unread badges and read cursors | |||||
| Do Not Disturb and quiet hours | |||||
| Scheduled send and reminders | |||||
| Status, presence and user groups | |||||
| Markdown, code blocks and link unfurling | |||||
| Rich cards and interactive blocks | |||||
| Command palette and keyboard shortcuts | |||||
| Install as an app | |||||
| Semantic search by meaning, not only keywords | |||||
| Message history | 90 days | Unlimited | Unlimited | Unlimited | Unlimited |
| Record video and screen clips | |||||
| Posting permissions per channel | · | ||||
| People from other organizations | 1:1 messages | Channels with up to 250 organizations | Channels with up to 250 organizations | Channels with up to 250 organizations | Channels with up to 250 organizations |
| Direct messages with people in other organizations With people you already share a channel with | |||||
| Chat apps and automation Wire chat to everything else | |||||
| Incoming webhooks | |||||
| Outgoing event subscriptions, signed | |||||
| Slash commands | |||||
| App integrations | Up to 10 | Unlimited | Unlimited | Unlimited | Unlimited |
| Interactive endpoints | |||||
| Recurring messages and standups | |||||
| Keyword automations | |||||
| Channel canvas one knowledge page per channel | · | Pinned page | Pinned page | Pinned page | Pinned page |
| Agents in chat Work started and finished in the conversation | |||||
| Agents in a channel | · | ||||
| Assistant DM | |||||
| Thread summaries, catch-up and daily recaps Runs on the organization's credit | |||||
| File summaries Runs on the organization's credit | |||||
| Chief of Staff Your assistant watches the workspace around the clock | · | Reports problems | Suggests fixes for you to approve | Suggests fixes you approve; fixing on its own coming soon | Suggests fixes you approve; fixing on its own coming soon |
| Proactive assistant and follow-ups | · | ||||
| Per-channel spend caps and alerts | · | ||||
| Chat secrets, encrypted | · | ||||
| Agent steering and cancel | · | ||||
| Link summarisation | · | ||||
| Charms Workflows on a canvas, run on a schedule or a trigger | |||||
| Charms per workspace a workspace limit, not per seat | 1 | 3 | 20 | 100 | Unlimited |
| Charm runs per month a workspace limit, not per seat | 50 | 500 | 2,000 | 20,000 | Unlimited |
| Visual canvas builder nodes, edges, pan and zoom | |||||
| Triggers: manual, schedule, webhook and event | |||||
| Cron schedule, to the minute | |||||
| Branching | |||||
| Filters on a step and on an edge | |||||
| Loops, with parallel iterations | |||||
| Retry with backoff, and timeouts | |||||
| Data mapping between steps | |||||
| Run history with each step's input and output | |||||
| Live run monitoring | |||||
| Step types agent, skill, connection, branch, filter, loop, wait, approval and more | |||||
| Web request, run code, web scrape and send email run code needs an operator allowlist; runtime cost is drawn from credit | |||||
| Emails sent by charms | 50/mo | Unlimited | Unlimited | Unlimited | Unlimited |
| Connections usable as a step runtime cost is drawn from credit | |||||
| Any OpenAPI schema as a connection | |||||
| Template library, install and publish | |||||
| Human approval, data input and AI review inbox | |||||
| Versioning, checkpoints and restore | |||||
| Export and import a blueprint | |||||
| Build a charm from a prompt runtime cost is drawn from credit | |||||
| AI steps: ask, extract, research, score, sort and filter runtime cost is drawn from credit | |||||
| Roles: viewer, runner, editor and owner | |||||
| Monthly budget cap on top of the charm run limit above | |||||
| Connections The systems your work already lives in | |||||
| Connection catalogue CRM, comms, CI/CD, databases, storage, HR, marketing and more | 200 | 200 | 200 | 200 | 200 |
| OAuth connections | |||||
| Incoming and outgoing webhooks | |||||
| Per-project encrypted credentials | · | ||||
| Connections as agent tools an agent uses a connection directly | · | ||||
| Bring your own OpenAPI schema any documented API becomes a connection | · | ||||
| Messaging bridges Slack, Teams and WhatsApp | · | · | |||
| Custom integration work | · | · | · | ||
| AI models and media Which model runs, and what it can produce | |||||
| Choice of AI model more than one model family, chosen per agent | · | ||||
| Reasoning level per agent low, medium, high or max | · | ||||
| Automatic model failover work continues when a model is unavailable | · | ||||
| Multi-agent reasoning agents that plan, hand work to each other and check the result | · | ||||
| Tool use and function calling | · | ||||
| Real-time web search | · | ||||
| Vision and image understanding | · | ||||
| Image generation | · | ||||
| Model choice per Plan & Work stage | · | · | · | ||
| Browser read | · | · | · | ||
| Video generation | · | · | · | ||
| AI model region choice choose where inference runs: US or China | · | · | · | · | |
| Voice Speak to it, and hear it back | |||||
| Voice dictation, commands and word list | · | ||||
| Agent replies read aloud | · | ||||
| Live voice conversation (US processing only) | · | ||||
| Voice notes in chat, Telegram and WhatsApp | |||||
| Spoken replies on Telegram and WhatsApp (on per connection) | · | ||||
| Meeting recording, speakers, summary and tickets | · | Audio kept 30 days | Audio kept 30 days | Audio kept 30 days | Audio kept 30 days |
| Phone number that takes a message | · | ||||
| Voice usage | · | Metered per minute in credits | Metered per minute in credits | Metered per minute in credits | Custom |
| Governance and compliance What an administrator controls | |||||
| Encryption in transit and at rest | |||||
| Retention policies, per channel | · | · | |||
| Audit log | Basic | Basic | Full | Full | Full |
| Data loss prevention | · | · | |||
| Compliance centre: export and legal hold legal hold, eDiscovery and data export | · | · | |||
| Single sign-on (OIDC) SAML and OIDC | |||||
| SAML single sign-on and SCIM provisioning | · | ||||
| Multiple identity providers and domain claiming | · | · | · | · | |
| Sign-in access logs | |||||
| See members' devices and sign them out | |||||
| Discovery API for exports | · | · | |||
| Information barriers | · | · | |||
| Data residency for files | · | · | |||
| HIPAA safeguards Technical controls; ask us for a BAA | · | · | |||
| Customer-managed encryption keys Add-on on Team and Premium: $150 per organization a month | · | · | Add-on | Add-on | |
| Custom roles | · | · | · | · | |
| Flagged-content review | · | · | · | · | |
| Anomaly alerts | · | · | · | · | |
| Self-hosted installation, with open-weight models inside | · | · | · | · | |
| Support Who answers, and how fast | |||||
| Support | Help centre | Priority | Priority | Dedicated + SLA | |
| First response | · | Within 4 hours | Within 4 hours, priority | Within 4 hours, priority | Per SLA |
| Email support | · | ||||
| Phone and video support arranged by email, then scheduled | · | · | |||
| Onboarding and implementation help we set the process up with you | · | · | · | ||
| Forward-deployed engineers our people working alongside yours | · | · | · | · | |
| Dedicated account contact | · | · | · | · | |
| 99.9% uptime SLA | · | · | · | · | |
Scroll the table sideways to see every plan.
Answers before you ask
No. Free does not expire and does not ask for a card. It is team chat, tickets, boards, pages and charms for up to three people on one project, with 90 days of message history and up to 10 app integrations. It includes 10,000 credits a month so you can try the assistant, summaries and AI charm steps, and you can buy more credit at 10,000 per $1 without upgrading. Runners and extra agents come with the paid plans.
No. We bill only members who were active in the last 28 days, and someone added part-way through a billing period is charged only for the rest of that period.
Yes. Nonprofits and schools can ask for discounted pricing from the billing settings, and we apply it once we have confirmed eligibility.
Yes. Upgrade or downgrade at any time from your billing settings. Upgrades take effect immediately with prorated charges. Downgrades take effect at the end of your current billing period. No data is lost when switching plans.
You are notified as the balance gets low. Your monthly allowance renews in full at the start of each billing period; if you use it all before then, you can top up at any time and work continues. Extra credits are 10,000 per $1, the same rate they are valued at when spent, with no bulk tiers, and topped-up credits never expire, so anything you do not use stays on the account. Your allowance is always spent before your purchased credits, so the pool that expires goes first. Projects, repositories and data stay accessible either way.
Yes. JuhJuh supports two-way sync with Jira, Linear, and GitHub Issues on all plans. Your existing issue tracker remains the source of truth. JuhJuh adds AI execution, diff review, and PR creation on top of your current workflow.
Where your rules ask for it, a person approves the spec before work starts, set by type, size, priority and labels, per project or per ticket. Each change is then built on its own isolated branch and checked against the ticket's requirements and acceptance criteria. After every step, JuhJuh checks the code inside a running copy of your services that mirrors production, then runs your tests and fixes what fails, before your pipeline and code review see it. A second agent on a different model reviews the change, your pipeline must pass, and a person decides every merge.
A credit is the single unit for everything the work consumes: AI model usage and the compute a run occupies. Model usage counts tokens in and out at the provider's published rate; compute counts the time a machine is actually working. Both convert into credits at 10,000 per $1, so one balance covers both and you never have to learn two units. Usage is tracked per ticket, per project and per person.
Every paid plan includes one computer of its own, always on, at a fixed address that does not change: 4 vCPU, 8 GB RAM and 160 GB of SSD on Starter, 8 vCPU, 16 GB and 320 GB on Team, 16 vCPU, 32 GB and 400 GB of NVMe on Premium. It is not stopped when idle and not wiped between sessions, so work you leave in progress is still there the next morning. The address is part of the machine and is not billed separately.
No, and it is worth saying plainly. You can stop, restart and resize the machine whenever you like, but a machine that is powered off still holds its disk and its address, so it still costs what it costs. Only destroying it would end the charge, and destroying it ends the product, so that one is handled by support rather than by a button.
Annual billing takes the same 16.67% off everything on the invoice and is billed once a year. The platform fee falls from $200 a month to $166.67 on Starter, $350 to $291.67 on Team and $900 to $750 on Premium, and the per-person price falls from $8.75 to $7.29 on Starter and $18 to $15 on Team and Premium. Your monthly credit allowance is unchanged and still renews every month.
Yes, on the Enterprise plan. Most teams start on our hosted service, which we run, update and support, with nothing to install. A team that needs more control, or must keep sensitive data inside, self-hosts JuhJuh in its own cloud account (AWS, Google Cloud, Microsoft Azure or DigitalOcean) or on its own premises, connected or air-gapped, with open-weight models from the Mistral, DeepSeek, GLM and Kimi families inside the boundary. Our engineers install it, tune it to your compliance obligations and move your workspace across when you're ready. You can also run both, each installation kept separate.
AI agents are autonomous assistants that execute multi-step tasks using 200+ tool integrations. They reason through problems, use tools, and produce results, all within the same approvals and safety gates as the rest of your workspace. The assistant is on every plan, including Free; more agents come with Starter plans and above.
Deployments push your application to infrastructure with one click. JuhJuh generates configuration from your project settings, manages versioning with semantic version bumps, and provides real-time logs. Secrets from your encrypted vault deploy atomically alongside your application. Available on Premium plans and above.
The vault stores secrets encrypted at rest. It auto-discovers environment variables from your configuration files, maintains snapshot history for rollback, and deploys secrets atomically to your infrastructure. Every change is versioned and diffable. Available on Team plans and above.
Start where you are. Scale when it earns it
Every plan ships with the five safety gates and a shared workspace for agents and people.