Let us host Helpin with AI usage included, or self-host and use your own providers. Invite your team without paying per seat. Agent features and limits vary by plan.
The complete open-source product, on your servers.
AGPL-3.0 · You cover hosting and AI providers
Community 0.2 is in beta. View on GitHub
Every module, hosted by us, with room to grow.
Per workspace · Billed monthly
For teams automating their process with custom agents.
Per workspace · Billed monthly
Self-hosting is free and complete. Cloud adds hosting, included AI, and support. Cloud plans start with a 14-day Growth trial, no card required. Prices in USD, taxes extra.
Compare all plansNeed a commercial license, a custom AI setup, or help with deployment? We’ll work through the requirements with your team.
Talk to us about EnterpriseRun agent tasks with the AI allowance included in your plan. You do not need separate provider keys to get started, and you control whether paid usage can go beyond the allowance.
Starter includes $99 of AI usage a month and Growth includes $299 ($79 and $239 on annual billing). The allowance resets monthly and doesn’t carry over.
Overage is off by default, so new AI work is declined until the allowance renews. Work already running keeps what it reserved. On an active paid plan, you can turn on metered overage at the same rates.
Self-hosted teams connect their own provider and pay it directly, with no Helpin AI fees. On Cloud, your own keys are available with Enterprise.
OpenAI
Anthropic
OpenRouter| Profile | Models | Input | Cache read | Cache write | Output / reasoning |
|---|---|---|---|---|---|
| SmallFast, efficient models for routine and high-volume work | GPT-5.6 Luna, GPT-6 Luna, GLM 5.3 Flash, DeepSeek V4.1 Flash, DeepSeek V4 Flash, Qwen3.8 Omni Flash | $0.33 | $0.033 | $0.275 | $1.32 |
| MediumStrong general-purpose models for broader reasoning and generation | Gemini 3.7 Flash, Gemini 3.8 Flash, GPT-5 mini, GPT-OSS 120B | $0.44 | $0.044 | $0.55 | $3.52 |
| LargeAdvanced models for planning, coding, review, and complex agent work | Claude Haiku 4.5, GPT-5.6 Terra | $2.20 | $0.22 | $2.75 | $13.20 |
| FlagshipHighest-capability models for the most demanding work | GPT-5.5, Claude Sonnet 5, Claude Opus 4.8 | $11.00 | $1.10 | $13.75 | $49.50 |
| Features and limits | Self-hostedFree · your serversSelf-host | Starter$99 / monthStart free | Growth$299 / monthStart free |
|---|---|---|---|
| Hosting | |||
| Hosting and updates | Your team | Helpin | Helpin |
| Your team | |||
| Users | Unlimited | Unlimited | Unlimited |
| Teams | Unlimited | 10 | Unlimited |
| Planning and delivery | |||
| Tasks & stories | Unlimited | Unlimited | Unlimited |
| Epics | Unlimited | Unlimited | Unlimited |
| Customer support | |||
| Round-robin assignment | Included | Not included | Included |
| AI conversation routing | Included | Not included | Included |
| Customer relationships | |||
| Contacts | Unlimited | 5,000 | Unlimited |
| Deal automation | Included | Not included | Included |
| Docs and help center | |||
| Documents | Unlimited | 500 | Unlimited |
| Multilingual help center | Included | Not included | Included |
| AI article translation | Included | Not included | Included |
| AI agents | |||
| Included AI usage per month | Your provider keys | $99 | $299 |
| Your own AI provider keys | Included | Not included | Not included |
| Custom agents | Included | Not included | Included |
| Agent scheduling | Included | Not included | Included |
| Extra AI usage | Paid to your provider | Metered, opt-in | Metered, opt-in |
| Automation and integrations | |||
| Automation flows | Included | Not included | Included |
| Branding | |||
| Widget branding | Removed | Helpin | Removed |
| Help and support | |||
| Support | Community | Standard | Priority |
Something else? Talk to our team →
Each workspace is its own subscription with its own billing. There is no per-seat charge, so adding teammates never changes the price.
No. The 14-day trial runs on Growth and includes $140 of AI usage. When it ends, the workspace is locked until you choose Starter or Growth.
New AI work is declined until the allowance renews next month; work already running keeps the allowance it reserved. On an active paid plan, you can turn on metered overage in billing settings to keep going at the same rates. Unused allowance doesn’t carry over, including on annual plans.
The model profile and the amount of work: input, cache, and output tokens, each at its profile’s rate. Billing settings show how much of the allowance you’ve used.
On a self-hosted install, always: connect OpenAI, Anthropic, OpenRouter, or a compatible endpoint and pay your provider directly. On Cloud, your own keys are part of Enterprise; Starter and Growth use the included allowance.
Yes. The open-source Community edition (0.2 beta) is free to self-host, with every module and no plan limits: support, projects, CRM, meetings, docs, automation, and AI agents, including coding agents. Your team runs the infrastructure and pays for it, along with any AI providers you connect.
Every product feature is open source under AGPL-3.0, including everything in Growth. The separate Enterprise license covers Helpin’s Cloud billing and hosted-AI metering code. Enterprise adds a commercial license for companies that can’t use AGPL, your own AI keys on Cloud, deployment help, and support terms.
Yes. The workspace’s billing owner can change the plan or billing period in billing settings. Check when the change takes effect and any price adjustment before you confirm.
Try the agents and automation in Growth with your team. No card required. Choose a plan when the trial ends.
Prefer your own servers? Self-host the open-source edition