Scale your agents.
Control your costs.

Start free. Pay one flat rate for the tokens your memory processes, and add fixed costs only when you add workspaces.

+ $0.10 per 1M tokens
  • 1 workspace included, more at $5 each per month
  • Unlimited users & API calls
  • Agentic integrations (Claude Code, Codex, MCP)
  • Data source integrations: Slack, Notion, Linear, Google Drive
  • In-app support
Popular

Everything in Standard, plus:

  • Auto-ontology: entity types learned from your own data
  • Ontology Editor: refine types and entities in plain words
  • Additional data domains to classify your data
  • Ontology kept up to date as your data grows
  • Priority support

Everything in Premium, plus:

  • Bi-temporal memory & conflict resolution
  • Provenance on every answer
  • Personalization per user & agent
  • Dedicated Slack channel
  • Dedicated support engineer
  • BYO cloud supported
  • Support SLA
Prefer your own infrastructure?

Cognee is open source. Run the full memory engine locally or on your own stack — free, forever.

View on GitHub →
Enterprise · BYOC engagements

Deploy the company brain in your own cloud.

The Enterprise plan is delivered as a fixed-scope BYOC engagement with our engineers: your ontology, evals on your data, and a runtime tuned to your domain, deployed in your VPC from day one. Pick the engagement length that fits, and if you're an early-stage startup, the Startup package gets you the same deployment at a fraction of the price.

StartupFor startups12 months · heavily discounted
BYOC deployment
Proprietary Cognee runtime
Context architecture / Ontology
Starter
Postgres adapter
Domain Pack / schema design + auto-generated schema
Auto-generated only
Customer-specific evals
Initial validation
Small-model runtime with our finetuned models
—
SSO / enterprise identity
—
Advanced permissions / governance
—
Architecture reviews
Kickoff
Support
Standard
6 months
BYOC deployment
Proprietary Cognee runtime
Context architecture / Ontology
Initial
Postgres adapter
Domain Pack / schema design + auto-generated schema
Initial
Customer-specific evals
Initial validation
Small-model runtime with our finetuned models
—
SSO / enterprise identity
—
Advanced permissions / governance
—
Architecture reviews
Monthly
Support
Standard
12 monthsRecommended
BYOC deployment
Proprietary Cognee runtime
Context architecture / Ontology
Ongoing
Postgres adapter
Domain Pack / schema design + auto-generated schema
Customer-specific evals
Automated / recurring
Small-model runtime with our finetuned models
Included
SSO / enterprise identity
Optional
Advanced permissions / governance
Limited
Architecture reviews
Bi-weekly initially
Support
Priority
24 months
BYOC deployment
Proprietary Cognee runtime
Context architecture / Ontology
Organization-wide
Postgres adapter
Domain Pack / schema design + auto-generated schema
Advanced
Customer-specific evals
Continuous + release gates
Small-model runtime with our finetuned models
Included + optimization
SSO / enterprise identity
Advanced permissions / governance
Architecture reviews
Ongoing strategic
Support
Named technical owner
Every engagement starts with evals on your own data.Book a callStartup pricing applies to pre–Series B companies.
Cloud runs on gpt-oss-120b — OpenAI's open-weight model
Get started

Cognee is the fastest way to start building reliable Al agent memory.