Pricing

Scale your LLM memory. Self-host free forever, or let us handle the infrastructure.

Free

$0/mo

For experimentation

  • 1 conversation
  • 50 requests / day
  • 2M token virtual context window
  • All 6 providers
  • Community support
Start Free

Pro

$19/mo

For production agents

  • Unlimited conversations
  • Unlimited requests
  • 10M token virtual context window
  • All 6 providers
  • Priority support
  • Session analytics
Start Pro Trial

Team

$99/mo

For teams and orgs

  • Everything in Pro
  • 100M token virtual context window
  • 5 team seats included
  • Shared memory namespaces
  • Dedicated support
  • Custom retention policies
Contact Sales

Enterprise

Custom

For regulated & on-prem

  • Everything in Team
  • Local installation support
  • Customization & white-label
  • Commercial licensing
  • SSO / SAML
  • Dedicated account manager
Contact Sales

How the managed window works

virtual-context is not selling a bigger raw prompt. It manages conversation state outside the live prompt, compresses it by topic, and pulls back only what matters when the model needs it. That is why a smaller managed window can outperform a larger full-history prompt: the active context stays curated instead of bloated.

The Free tier is designed for evaluation, local prototypes, and solo experiments. Pro is the default for production agent work where you need unlimited requests, longer memory horizons, and fewer operational decisions. Team is built for multi-user rollouts, shared memory needs, and larger managed workloads.

Hosted or self-hosted

Every plan includes the open source core under AGPL-3.0, so you can self-host and keep full control of your infrastructure. The managed cloud product adds tenant setup, billing, dashboard tooling, and hosted operations around the same engine. Teams often start by self-hosting during evaluation, then move to the managed product when they want faster rollout or less infrastructure work.

Cloud subscriptions are billed monthly. Upgrades and cancellations are handled in the dashboard billing flow, and cancellations remain active through the end of the current billing period.

Pricing FAQ

What does the virtual context window mean on each plan?

The virtual context window is the budget virtual-context manages on your behalf after compaction, retrieval, and assembly. It is not a promise that every request sends millions of raw tokens to the model. Instead, the system keeps a much smaller active prompt while preserving recoverable memory outside the prompt. Free is designed for experiments and small workloads, Pro is the default for production agent sessions, and Team is meant for shared or high-volume deployments that need a much larger managed memory budget.

When should I upgrade from Free to Pro or Team?

Upgrade to Pro when you move from evaluation into real day-to-day agent work, need unlimited requests, or want longer-lived conversations without aggressively trimming history. Team is the right fit when you need shared usage across multiple people, larger managed context budgets, or coordination around support, retention, and rollout planning. If you are self-hosting and only need the open source engine, the AGPL core remains available without a cloud subscription.

What is the difference between the hosted product and the open source core?

The hosted product bundles the same memory engine with managed infrastructure, billing, account controls, tenant provisioning, and the dashboard. The open source core is ideal when you want to self-host the proxy, storage, and runtime yourself. Both paths use the same core concepts: topic-aware compaction, retrieval, structured facts, and managed context assembly. You can evaluate locally, then move to the hosted service when you want less operational overhead.

How does billing and cancellation work for cloud plans?

Cloud plans are billed monthly. You can upgrade from the billing area in the dashboard, and cancellations stay active through the end of the current billing period. The Free tier is intended for experimentation, while paid tiers are for sustained production use. Team and enterprise conversations can also start through Contact Sales if you need procurement support, rollout planning, or contract review before enabling a managed deployment.