Simple pricing. No surprises.
Every plan includes a fully hosted agent on a dedicated server, Active Memory, automatic updates, and the full onboarding experience. Pick the plan that fits your team.
Individual
14-day free trial — no credit card required
For solo operators and individual contributors.
- 1 AI agent on a dedicated server
- Full Active Memory system (4-tier graduation)
- AI-guided onboarding and knowledge hub
- Automatic runtime updates
- BYOK — bring your own LLM API key
- Community support
Teams
14-day free trial — no credit card required
For teams that need shared context across agents.
- Everything in Individual
- 3 AI agents on dedicated servers
- Multiple communication channels
- Shared team-level memory
- Knowledge graduation across agents
- Priority support (email, 24-hour response)
Enterprise
starting at
5 agents included · +$49/mo per additional agent
For organizations that need the full memory hierarchy.
- Everything in Teams
- Full organizational memory
- Automatic knowledge promotion across teams
- Dedicated memory stack
- Custom onboarding configuration
- Dedicated support (Slack, 4-hour response)
- SSO and advanced access controls (roadmap)
Plan comparison
| Feature | Individual | Teams | Enterprise |
|---|---|---|---|
| Agents | 1 | 3 | 5 (+ add-ons) |
| Dedicated servers | ✓ | ✓ | ✓ |
| Channels per agent | 1 | Multiple | Multiple |
| Active Memory | ✓ | ✓ | ✓ |
| 4-tier knowledge graduation | ✓ | ✓ | ✓ |
| Agent-scope memory | ✓ | ✓ | ✓ |
| Team-scope memory | — | ✓ | ✓ |
| Company-scope memory | — | — | ✓ |
| Cross-agent knowledge sharing | — | ✓ | ✓ |
| Cross-team knowledge promotion | — | — | ✓ |
| BYOK security | ✓ | ✓ | ✓ |
| Vault encryption | ✓ | ✓ | ✓ |
| Automatic updates | ✓ | ✓ | ✓ |
| Onboarding interview | Standard | Standard | Custom |
| Support | Community | Priority (email) | Dedicated (Slack) |
| SSO | — | — | Roadmap |
Frequently asked questions
Do I need to provide my own LLM API key?
Yes. Cortex uses a Bring Your Own Key (BYOK) model. You provide your Anthropic, OpenAI, or Moonshot API key. This means you control your LLM costs directly and can choose the model that works best for your use case. Your keys are encrypted in Supabase Vault and never stored on the agent server.
What does "dedicated server" mean?
Every Cortex agent runs on its own isolated cloud server (Ubuntu, 2vCPU/4GB RAM). This isn't a shared container — your agent has its own filesystem, processes, memory stack, and SSL certificates. No other customer's data touches your infrastructure.
How long does setup take?
About 10 minutes. You walk through an AI-guided knowledge interview about your business, connect your channel, and provide your API key. Cortex provisions the full infrastructure in the background while you're still onboarding.
Can I upgrade or downgrade?
Yes. You can upgrade at any time and your agents' memory carries over. Downgrading works similarly — just reach out to support.
What if I need more than 5 agents?
Enterprise starts with 5 agents at $1,999/mo. Every additional agent is $49/mo. Contact our sales team if you need a custom arrangement.
What LLM costs should I expect?
LLM costs depend on your usage and chosen provider. A typical agent using Claude Haiku for memory extraction and a more capable model for responses might cost $20–100/month in API fees, depending on conversation volume. This is billed directly by your LLM provider, not by Cortex.
Is there a free trial?
Yes. Every plan includes a 14-day free trial — no credit card required. You can spin up an agent, connect it to Slack or Telegram, and see it live in under 10 minutes.
What happens to my data if I cancel?
Your agent's memory and knowledge base can be exported before cancellation. After your subscription ends, servers are deprovisioned and data is deleted within 30 days.
Ready to deploy your first AI teammate?
Get set up in 10 minutes. No DevOps required.