Pricing

Simple, transparent pricing

Last reviewed: 2026-05-31

Start free. Scale as your agents go to production. No hidden fees.

Klio is one shared brain for all your AI agents — connect Claude, Cursor, Codex, and any MCP agent to a single shared memory, so what one learns, they all know. Your agents stop forgetting, repeating work, and contradicting each other. Local-first, encrypted, and open source.

  • Free

    $0/mo

    Start sharing memory across your agents.

    Get Started Free
    • Memories captured1,000 / mo
    • Recalls10,000 / mo
    • Cross-agent syncLocal only
    • Connected agentsUnlimited
    • Memory retention1 day
    • Rate limit100 RPM
    • OverageHard limit
    • SupportCommunity
  • Starter

    $29/mo

    For solo devs running a few agents together.

    Start Starter
    • Memories captured25,000 / mo
    • Recalls100,000 / mo
    • Cross-agent syncReal-time
    • Connected agentsUnlimited
    • Memory retention7 days
    • Rate limit500 RPM
    • OverageHard limit
    • SupportEmail
  • Most Popular

    Pro

    $99/mo

    or $990/yr ($83/mo) billed annually

    For teams whose agents work together daily.

    Start Pro
    • Memories captured150,000 / mo
    • Recalls1,000,000 / mo
    • Cross-agent syncReal-time
    • Connected agentsUnlimited
    • Memory retention30 days
    • Rate limit1,000 RPM
    • Overage$0.0005 / memory, $0.0001 / recall
    • SupportEmail (48h)
  • Team

    $349/mo

    or $3,490/yr ($291/mo) billed annually

    For organizations running many agents at scale.

    Start Team
    • Memories captured1,500,000 / mo
    • Recalls10,000,000 / mo
    • Cross-agent syncReal-time + priority
    • Connected agentsUnlimited
    • Memory retention90 days
    • Rate limit5,000 RPM
    • Overage$0.0004 / memory, $0.00008 / recall
    • SupportPriority (24h)

Enterprise

Need custom limits, SLAs, SSO, or on-prem deployment? Let's talk.

Contact Sales

How Vex Compares

See how Vex stacks up against other AI agent reliability tools.

Supported~ Partial$ Paid only Not available
Vex vs Guardrails AI feature comparison
FeatureVexGuardrails
Runtime Verification
Hallucination detection~
Behavioral drift detection
Schema validation
Multi-turn coherence
Sync verify (real-time)
Auto-Correction
Auto-correction cascade~
Graduated correction layers
Transparent / opaque modes
Observability
Async telemetry ingestion
LLM call tracing
Cost & token tracking
Dashboard & analytics$
Developer Experience
Python SDK
TypeScript SDK
Setup complexity3 linesModerate
Framework agnostic
Pricing & Plans
Free tier
Self-hosted option
Full Vex vs Guardrails AI breakdown →

Frequently Asked Questions

What counts as an observation?

An observation is any LLM call, tool invocation, or agent action that Vex monitors. Each individual event in your agent pipeline counts as one observation.

What counts as a verification?

A verification occurs when Vex actively checks an agent output against your defined policies, behavioral baselines, or factual constraints. Verifications consume more compute than passive observations.

What happens when I exceed my plan limits?

On Free and Starter plans, monitoring pauses until the next billing cycle. On Pro and Team plans, you can continue beyond your included quota at the listed overage rates. You will receive alerts as you approach your limits.

Can I switch plans at any time?

Yes. You can upgrade or downgrade your plan at any time. Upgrades take effect immediately with prorated billing. Downgrades take effect at the start of your next billing cycle.

Do you offer annual billing?

Yes. Annual billing saves you roughly two months compared to monthly pricing. Pro is $990/yr ($83/mo) and Team is $3,490/yr ($291/mo). Contact us to switch to annual billing.