← Back to Codebase Memory: MCP for AI Agents

Simple, Predictable Pricing

Reduce your AI context tokens by 99% without breaking the bank. Choose the plan that scales with your codebase.

Starter

Free

For small projects and experimentation

Enterprise

Custom

For large organizations and specialized needs

Frequently Asked Questions

What counts toward indexed storage?

Indexed storage includes the compressed codebase snapshots and metadata. We index only code files; binaries, node_modules, and build artifacts are excluded by default.

Can I cancel anytime?

Yes. Pro plans are month-to-month with no lock-in. You can downgrade or cancel at the end of your billing cycle without penalty.

Do you offer annual discounts?

Yes. Pro plans billed annually receive 20% off. Contact sales@codebase-memory.dev for bulk discounts and multi-year pricing.

What's the difference between memory extraction modes?

Starter includes standard symbol extraction. Pro adds context-aware relationships, cross-file dependencies, and architecture patterns. Enterprise includes custom extraction plugins.

Is there a free tier for open source?

Yes. Open source projects on GitHub/GitLab qualify for a free Pro tier upgrade. Email us with your repo link.

How does the token reduction work?

Codebase Memory creates a compressed index of your code structure. Instead of passing entire files to Claude, you pass the index (typically 1% of original size), then reference specific code sections by identifier.