Simple monthly seat. Token credits priced with a transparent markup over the underlying model cost — every dollar visible in your dashboard, per agent, per work item, per minute.
Token usage is metered separately and billed monthly against your credit balance. Top up any time; unused credits roll over.
Every agent call consumes tokens. The rate below is what you pay — a transparent markup over our underlying model cost that funds the platform around it. You see the exact spend per agent, per work item, per minute in your dashboard.
| Model | Best for | Input $ / 1M tokens | Output $ / 1M tokens |
|---|---|---|---|
Sonnet 4.5 Balanced | Default. Good for most agent work. | $18.00 | $90.00 |
Opus 4.5 Deep thinking | Long reasoning, complex planning, code. | $90.00 | $450.00 |
GPT Image 1 (OpenAI) Fast | Quick replies, triage, simple lookups. | $0.00 | $0.00 |
GPT-4o mini TTS (OpenAI) Fast | Quick replies, triage, simple lookups. | $3.60 | $0.00 |
Gemini 2.5 Flash Image Fast | Quick replies, triage, simple lookups. | $0.00 | $0.00 |
Haiku 4.5 |
Each user with edit access to your org is $25/month. Viewers are free. Add or remove anytime — we prorate.
Every agent call records exact tokens, exact cost. You see it live in your dashboard, broken down by team / agent / work item.
Buy credits in advance. When an agent runs, credits get drawn down at the rate shown above. Top up anytime — unused balance never expires.
You can. But you'd be writing your own agent runtime, work tracker, scheduler, integrations layer, and team-management UI. We've done that. The token markup pays for the platform that turns tokens into a team.
Agents pause until you top up — they don't fail silently. Scheduled work items queue. You get an email at 80% and at 0%.
Yes. Viewer-role users (read-only access) don't count toward your seat count. Only admins and members are billable.
Yes — full JSON export of every team, agent, work item, and conversation, anytime, from the Settings → Export / Import tab. No lock-in.
| Quick replies, triage, simple lookups. |
| $6.00 |
| $30.00 |
TTS-1 (OpenAI) Fast | Quick replies, triage, simple lookups. | $90.00 | $0.00 |
TTS-1 HD (OpenAI) Fast | Quick replies, triage, simple lookups. | $180.00 | $0.00 |
text-embedding-3-large (OpenAI) Fast | Quick replies, triage, simple lookups. | $0.78 | $0.00 |
text-embedding-3-small (OpenAI) Fast | Quick replies, triage, simple lookups. | $0.12 | $0.00 |
Prices in USD, per 1,000,000 tokens. Cached input is billed at the same rate as a cache read on supported models. Your usage rolls up in your billing dashboard.