AI credits — how they work
Every plan includes a monthly pool of AI credits; you only spend them when the agent does real work.
AI Emaily meters AI in simple, usage-based credits. Every plan comes with a monthly pool, and a credit is only spent when the agent does real work — summarising a thread, drafting a reply, or acting for you. Reading, syncing, search, and navigating never cost credits.
What a credit is
A credit is the unit of AI work. Different actions cost different amounts, so light tasks barely touch your pool and only full autonomous actions cost more.
| Action | Credits | What happens |
|---|---|---|
| Thread summary | 0.5 | The agent condenses a whole conversation to what matters. |
| Drafted reply | 1 | A full reply written in your voice, grounded in your context. |
| Autopilot reply | 2 | Summarise, draft, and send — hands-free, within your rules. |
What each plan includes
Credits refresh at the start of each billing cycle. Team credits are allotted per seat and pool across the whole team, so heavy users can draw on quieter teammates’ allotments.
| Plan | Monthly credits | What that gets you |
|---|---|---|
| Free | 10 | ≈ 20 summaries, or 10 drafts to try the agent. |
| Pro | 500 | ≈ 500 voice-matched drafts, or 1,000 summaries a month. |
| Autopilot | 1,000 | ≈ 500 fully autonomous replies, sent for you. |
| Team | 1,000 / seat | Full Autopilot credits per seat, pooled across the team. |
Run unlimited with your own key (BYOK)
On any paid plan you can bring your own Anthropic, OpenAI, or Google key. When BYOK is on, credits don’t apply — you run unlimited AI and pay your model provider directly for usage. It’s the recommended path for heavy users and lifetime-deal buyers.
No caps with BYOK
What happens if I run out?
Nothing breaks. The full email client keeps working — only AI actions pause until your credits refresh next cycle. You can upgrade your plan for a larger pool, or switch on BYOK to remove the limit immediately.
Frequently asked
Ready to try it?
Start free