Free: up to 10 team members, 1 page, 50 facts. Small Teams: €9/month (4 team members included, +€3 per extra), 200 pages, 5,000 facts. Large Teams: €49/month (20 team members included, +€2 per extra), unlimited pages, 50,000 facts. Yearly billing available at €90/year and €490/year.
Key features
Harpe: chooses what an AI agent reads under a token budget with consistent results
Kibisis: keeps agent memory on your hardware with AES-256-GCM encryption
Argus: append-only record that detects edits with cryptographic verification
No network code by default for local and disconnected use
MIT License with published code, benchmarks and raw results
Shared fact management across team AI tools
Granular read and write permissions per page
Export as Markdown or JSON
Per-workspace encryption
Works with Claude, ChatGPT, Cursor, Copilot and MCP-compatible tools
No installation required, connector-based setup
API support for automated exports
Sealed pages unreadable even to Brainpot staff
What makes it different
Measured empirical validation across 6,111 questions outperforming full-context models
Zero plaintext on disk with encrypted vaults matching plaintext accuracy
78.3% fewer tokens sent to the model with a 1,200-token budget per turn
Preserves conversational chronology and relational qualifiers under token budgets
No token costs; Brainpot runs no AI and integrates with existing tools
Full data control with 90-day grace period after cancellation
Perseus Computing